ad aan, A ng Man ar ve ! = Be . : ; B DEE S Wr N; ‘ NE Baer RER REN n: BER, u ae eraadız Eine Regine Es x Sp Taete [3 ser ER ne tu Arie Na A Re ade Ferne = “ Han KERTRE E EUR Sr: hal, RN: wAlge as HN in *3 are TOR Y Bine EIER AUS BZ TRERELIG we engere ee en Sn u ER 7 SE en BERNIE Te ag Re Ana a RS Da re kazager: ie > < > = = = = > = zZ = = T . 5 = ö z 5 Zi E 2, = = 2 5 | Be = a = 177) JILNLILSNI_NVINOSHLINS S3IUVY4AIT LIBRARIES SMITHSONIAN INSTITUTION a n = % Z Sr {N us in ul 7 „4 4 re = u Ber = c De c = e = De = R= 4 = 5 = = m» ö = = RN Hr = => nn BRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31IUVYY4917 LIBRAR ö - = = z E - 3 : E - = au ad = r) = =.) > > — > _ > > = = = a pm % un N Bas N z o z 2 N oO ILNLILSNI S3I9VY8I1_LIBRARIES _SMITHSONIAN 7 “nr — ns: “N Zr 2) = =, 3 N = 5 2 = = E = 3 2 9 f7 = en er = = = Si... i 177) = 177) = ui 177) ne z BRARIES SMITHSONIAN INSTITUTION NOILNLILSNI S314vY4gı7 LIBRARF an zZ n > N z \ & .ö = 2 u GR zs = = 3 = = a” S oc = x S © | 5 nn N 5 2 o ‚DC Ber Ü = _ e) = eu = [| NLNLILSNI _NVINOSHLIWS 53 Iyuvyg 1_ LIBRARI ES, SMITHSONIAN | INSTITUTIONEERESESTEEE Er oO = Ja Ne oO DEN oO = = % Ze ., 7 E = > ; > = er E = > a > 2m ee RN B= ES ee > BRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31uvuaıT _LIBRARE = Ss SSA0TS < 2 n > P 5 < I \ = ı\ z - = DET: N N 2 \ = 2, zy ae 2 nes wre, 8 > 5 N er = in BI > = 197) Sr ILNLILSNI NVINOSHLINS S3IUWNH17 LIBRARIES NOILNLIN 2 BD = u z o i ” a 7) 2 77) = == x = x se RG = .< =. < ie < = ce = x Bar x ° 2 @ 65 Br ö = = = z er zZ ir | BRARIES SMITHSONIAN INSTITUTION NYINOSHLINS S313YY811 LIBRAR ö - = = z Ze = a I ee) = E r % > 25 > > > © m = = a un $> 2) we je: n- mE n „2 IILNLILS x „sata vun! „U! BRARIES, SMITHSONIAN INSTITUTIOWES < = > „= SS .. = ; < N \ = | v4 = + LG G z N == zZ N g en IE 4 E € 70 = = E 777, = 2 » 77) Lu M BRARIES SMITHSONIAN _ INSTITUTION NOILNLILSNI NVINOSHAIWS S3IUV4MSI1_LIBRARF m ren rn N nn. = u ıe 72) TEE rd h 5 & S = = = = e z = N = z - L 'e) T. ke) =E (e) \ I N [7] IR, 0 (22) (9) NR (0%) > T ) Bes oO 2.2 N [®) : = z, = = ZEN € = = N >" = == = IS > = 172) Nr z 177) 4 77) = = _NVINOSHLINS S31UVYYEAIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_NVIN Zi. RE ie zZ n = 0 zn N u (97) ul [9 u N R A & je © 5 ZI x a 5, m Wr. = < ad < In 3 RN 04 S x . 5 X S Sen 2 =) a) ö 2 fe) = — = er z En zZ SMITHSONIAN INSTITUTION NOILNALILSNI NVINOS S31ISNY49I71 LIBRARIES SMII : - 3 = E - 2 = FE: = 2 = 2 = z n) 2720 = 3 5 > E; - DL; = > = m ® m ° Zu z ” = D 2 NVINOSHLINS S3IUMYYMAIT LIBRARIES SMITHSONIAN NOILNALILSNI Be: (N zZ. (2 —e 7) 2x - = KL oe = < S „= z I sh 5 7 zZ = 09 2 3 DH ° 2 2 3,74 ? = 7 8: = E = pP = > z = > = > . KH = n z n <= 7) >= 77) SMITHSONIAN INSTITUTION LIBRARIES SMIT n zZ an 3 (75) = (02) u (77) vu 7) = un 2 r. —_ TE —_ = — TE en. — x _ x _ Be. T; = = = T = = dd c jur Jar x = x 7, e) = je = m = me -_ Ach \®) na i®) — oO nn J 4 | = ii Ba 2 PR _NVINOSHLINS_S31UV4g171_LIBRARIES SMITHSONIAN _INSTITUTION NOILNLILSNI_NVIN .%“ 2 ® = s _ 5 ZEN D = n [0 = Eu D 5 = D ae, E an > = >» r > r NEN, 4 = ud rn, > > ZONE. n = - n m 2 m | n = u = 72) = 72) SMITHSONIAN INSTITUTION NOILNLILSNI NYVINOSHLINS LIBRARIES SMIT a Ei n zZ un ey: Q a < = < = = = a. = iur) = ar = = E (e) RN TE 2 © 2 @) T [dp] in N - (N Ri (dp) (N? (N N B] T EI e) T ®) I ®) = = Nr E = = = > = N >" = > => N = 72) le z 77) > 77, g: = _NVINOSHLINS S31UVY81I7 LIBRARIES INSTITUTION NOILNLILSNI _NVIN N ER N ul (nn u N 2 . X = x ? „I x = h = NO < = < N < N e. NR m 4 = zn = = : Ws De ° = ° - = z BR = = = En, = -SMITHSONIAN INSTITUTION NOILNALILSNI NVINOSHLINS S3IUVYM9I7 LIBRARIES SMI = ER z m z r zZ e ; ®) = oO Se oO 5 / PA Rt = Z 2 r 2 E ? : 2% Dj 4 >.) /; 027, > = = = z 37. | 7, 2 m 2 > > > 7 WR m = - 2 m u 3 — [07] 2 == (N == NVINOSHLIWS S3 I4avua N_Lı BRAR!I ES „SMITHSONIAN INSTITUTION ., NOILNLILSNI NYIN z = a = 2,0. = ” = (= =) z SS 3 z 2 I. QO.« = ©) SON => (®) 2 Er = 3 ER 3; 2 3 = = = = z = E >» 2 = Sul ee = > = N 2 (7) = n er [7) SMITHSONIAN _ INSTITUTION NOILNLILSNI NVINOSHLUNS S3IYYYH11_ LIBRARIES SMIT n ze en 0 BETEN => PILT oe n nn => mL ) ZEITSCHRIFT FÜR SAÄUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Organ der Deutschen Gesellschaft für Säugetierkunde Volume 53, 1988 ISSN 0044-3468 Herausgeber / Editors P. J. H. van Bree, Amsterdam — W. Fiedler, Wien - H. Frick, München —- W. Herre, Kiel - H.-G. Klös, Berlin — H.-J. Kuhn, Göttingen - E. Kulzer, Tübingen - B. Lanza, Florenz — J. Niethammer, Bonn — H. Reichstein, Kiel — M. Röhrs, Hannover — D. Starck, Frankfurt a. M. - F. Strauß, Bern - E. Thenius, Wien - P. Vogel, Lausanne Schriftleitung/Editorial Office H. Schliemann, Hamburg - D. Kruska, Kiel Mit 176 Abbildungen Verlag Paul Parey Hamburg und Berlin Wissenschaftliche Originalarbeiten AHLEN, 1.: Sonar used by flyıng Lesser horseshoe bats, Rhinolophus hipposideros (Bechstein, 1800) (Rhinolophidae, Chiroptera), in hunting habitats. - Ortungslaute von fliegenden Kleinen Hufeisennasen, Rhinolophus hipposideros (Bechstein, 1800) (Rhinolophidae, Chi- roptera), inJagdbiotepen.. .& ... we ».mmge.. „ ar. Me ee. 2 ARLETTAZ, R.; AULAGNIER, $.: Statut de trois especes de chiropteres rares au Maroc: Nycteris thebaica, Hipposideros caffer et Pipistrellus rueppelli. — Status of three rare bat species in Morocco: Nycteris thebaica, Hipposideros caffer and Pıpistrellus rueppelli. - Zum Vorkom- men von drei seltenen Fledermausarten in Marokko: Nycteris thebaica, Hipposideros caffer und'Pıpistrellustueppel, 2 .... » Deus ee a BAKER, CAROLYN M.: Vocalizations of captıve Water mongooses, Atılax paludinosus. — Vokali- sationen des Sumpfichneumons, Atılax paludinosus, in Gefangenschaft ............ BAKER, CAROLYN M.: Scent marking behaviour in captive Water mongooses (Atilax paludino- sus). — Duftmarkierung bei gefangengehaltenen Sumpfichneumons (Atzlax paludinosus). . . . BALAKRISHNAN, M.: Structure of Lepus nigricollis hair from various body regions with Scanning Electron Microscopy. — Struktur der Haare von Lepus nigricollis aus verschiedenen Körper- resionen imirasterelektronenmikroskopischen Bilder 2 2 Bastian, H. V.: Vorkommen und Zug der Rauhhautfledermaus (Pipistrellus nathusii Keyserling und Blasius, 1839) in Baden-Württemberg. — The occurrence and migration of Nathusius’ pipistrelle (Pipistrellus nathusii Keyserling & Blasius, 1839) in Baden-Württemberg. .. .... LE BERRE, M.; LE GUELTE, L.: Structure de l’espace et retour au nid chez la gerbille de Mongolie (Meriones ungniculatus). — Space utilization and homing in the Mongolıan gerbil (Meriones ungniculatus). — Raumnutzung und Rückkehr zum Nest bei der Mongolischen Rennmaus (Meriones unguiculatus) . . =. ». = nenne se ee Broop, B. R.; McFarLANE, D. A.: Notes on some bats from northern Thailand, with comments on the subgeneric status of Myotis altarınm. — Über einige Fledermäuse aus dem nördlichen Thailand mit Bemerkungen zur subgenerischen Zugehörigkeit von Myotis al- ara... nee ee Bock, W. F.: Die Bedeutung des Untergrundes für die Größe von Bauen des Dachses (Meles meles) am Beispiel zweier Gebiete Südostbayerns. — The importance of soil composition for the sıze of badger setts (Meles meles), ıntwoareasinSE Germany .... 2... 2.2.2.2... BoLLes, KATHRYN: Evolution and variation of antipredator vocalisations of Antelope squirrels, Ammospermophilus (Rodentia: Sciuridae). - Zur Evolution und Variation der Warnrufe der nordamerikanischen Zieselgattung Ammospermophilus (Rodenua: Sciuridae) ......... BovEr, ]J.; DoLivo, M.; GEORGE, C.; GOGNIAT, A.: Homing behavior of Wood mice (Apode- mus) ın a geomagnetic anomaly. — Heimfindeverhalten von Wald- und Gelbhalsmäusen (Apodemus) ın einer geomagnetischen Anomalie 2 2... 2 BowLann, A. E.; PERRINn, M. R.: The effect of fire on the small mammal community ın Hluhluwe Game Reserve. — Der Einfluß von Bränden auf die Kleinsäuger im Hluhluwe- Wildreservat . 2... 2 ua ES a a ee a Sn ee BRUORTON, M. R.; PERRIN, M. R.: The anatomy of the stomach and caecum of the Samango monkey, Cercopithecus mitis erythrarchus Peters, 1852. — Die Anatomie von Magen und Blinddarm der Diadem-Meerkatze Cercopithecus mitis erythrarchus Peters, 1852 ....... DARDAILLON, MaARYSsE: Wild boar social groupings and their seasonal changes in the Camargue, southern France. — Soziale Gruppenbildungen und ihre jahreszeitlichen Änderungen bei Wildschweinen in der Camargue, Sudirankreich. 2.2 2. 22 222 DusosT, G.; FEER, F.: Variabilite comportementale & l’interieur du genre Cephalophus (Rumi- nantia, Bovidae), par l’exemple de C. rufılatus Gray, 1846. — Verhaltensunterschiede innerhalb der Gattung Cephalophus (Ruminantia, Bovidae) am Beispiel von C. rufllatus . . . FERRARI, C.; Rosst, G.; Cavanı, C.: Summer food habits and qualıty of female, kid and subadult Apennine chamois, Rupicapra pyrenaica ornata Neumann, 1899 (Artiodactyla, Bovidae). — Beschaffenheit und Qualität der Sommernahrung von Weibchen, Kitzen und der Apenningemse, Rupicapra pyrenaica ornata Neumann, 1899 (Artiodactyla, Bovidae) ..2 4.2.2... N ar N FIsCHER, M. $.: Zur Anatomie des Gehörorganes der Seekuh (Trichechus manatus L.), (Mam- malıa: Sirenia). — Contributions to the anatomy of the hearing organ of the seacow (Iinchechussmanatus 1..), (Mammala: Sirema)ı. 2. FuURLEY, CH. W.; Tıchy, H.; UERPMAnN, H.-P.: Systematics and chromosomes of the Indian gazelle, Gazella bennetti (Sykes, 1831). — Systematik und Chromosomen der Indischen Gazelle, Gazellaibennett: (Sykes, 1831) 22282 GISBERT, J.; LOPEZ-FUSTER, M. ]J.; GArRcfa-PEREA, Rosa; VENTURA, ]J.: Distribution and biometry of Sorex granarins (Miller, 1910) (Soricinae: Insectivora). — Verbreitung und Biometrie von Sorex granarius (Miller, 1910) Soricinae: Insectivora) .............. 65 321 83 358 69 202 225 276 349 129 333 235 210 DD: 31 170 365 48 GRANJON, L.; CHEYLAN, G.: Me&canismes de coexistence dans une guilde de murides insulaires (Rattus rattus L., Apodemus sylvaticus L. et Mus musculus domesticus Rutty) en Corse: Consequences Evolutives. - Mechanisms of coexistence in a guild of insular Murids (Rattus rattus L., Apodemus sylvaticus L. and Mus musculus domesticus Rutty) ın Corsica: Evolutio- nary consequences. - Mechanismen der Koexistenz von Insel-Muriden (Rattus rattus L., Apodemus sylvaticus L. und Mus musculus domesticus Rutty) auf Korsika: Evolutionäre IKRorSsequen Ze ne ee ers: BE ende A Hansson, L.: Parent-offspring correlations for growth and reproduction ın the vole Clethriono- mys glareolus in relation to the Chitty Hypothesis. -— Korrelationen von Wachstum und Reproduktion zwischen Eltern und Jungtieren bei der Rötelmaus Clethrionomys glareolus ın Bezichungzur Cinszlubyodıese ee oe ee ee HERTENSTEIN, BIRGIT; ZIMMERMANN, ELKE; RAHMANN, H.: The development of visual acuity in treeshrews (Tupaia belangeri). - Die Entwicklung der Sehschärfe bei Spitzhörnchen BorpaAbelangeri)) Anne 0 ru a hama ae ee rn Herzog, S.: The karyotype of the European roe deer (Capreolus capreolus L.). - Der Karyotyp desfennopäischensRehes(@apreolusicapreolusle) 2 ee Hıckman, G. C.: The swimming ability of Ctenomys fulvus (Ctenomyidae) and Spalacopus cyanus (Octodontidae), with reference to swimming ın other subterranean mammals. — Die Schwimmfähigkeit von Ctenomys fulous (Ctenomyidae) und Spalacopus cyanus (Octodonti- dasyyima\lereleichizuranderen subterranen Saugetieren ne. 2 22 000. yon JOHANNESSON-GROSsS, Krıstina: Lernversuche in einer Zweifachwahlapparatur zum Hell- Dunkel-Sehen des Maulwurfs (Talpa europaea L.). — Brightness discrimination of the mole (Talpa europaea L.) in learning experiments applying amodified tube-maze method ..... Könıg, A.; ROTHE, H.; SıEss, MARGARETHA; DARMS, K.; GRÖGER, DAGMAR; RADESPIEL, UTE; Rock, J.: Reproductive reorganızation in incomplete groups of the common marmoset (Callithrix jacchus) under laboratory conditions. — Reproduktive Reorganisation in unvoll- ständigen Gruppen des Weißßbüscheläffchens (Callthrix jacchus) unter Laborbedingungen . . MıLkovsk$, ]J.: Secondary sex ratio in the Przewalski horse Equus przewalski (Mammalıa: Equidae). — Das sekundäre Geschlechterverhältnis beim Przewalski-Pferd Equus przewalsku NNlemmalie: Beudae) ao als a es ee MÜLLER, E. F.; Soppa, U.: Activity pattern and thermoregulation in the Cuis (Galea musteloides Meyen, 1833). - Aktivitätsmuster und Temperaturregulation beim Wieselmeerschweinchen (Galea mnstelaoaes)N ara, leSs)E PATTERSON, I. J.: Responses of Apennine chamois to human disturbance. — Reaktionen aBenamıschen@emsen autmenschlichesStorung.. 2 ... nu... nun. PETERS, J.: Osteomorphological features of the appendicular skeleton of African buffalo, Syncerus caffer (Sparrman, 1779) and of domestic cattle, Bos primigenius f. taurus Bojanus, 1827. — Osteomorphologische Unterscheidungsmerkmale am Gliedmaßenskelett vom afrı- kanıschen Büffel (Syncerus caffer) und vom Hausrind (Bos primigenins f.taurus) ....... ROTHE, H.; Könıc, A.; RADESPIEL, UTE; DARMS, K.; SIESS, MARGARETHA: ÖOccurrence and frequency of twin-fight in the Common marmoset (Callithrix jacchus). — Auftreten und Häufigkeit von Zwillingskämpfen beim Weißbüscheläffchen (Calhthrix jacchus) ....... STORCH, G.: Eine jungpleistozäne/altholozäne Nager-Abfolge von Antalya, SW Anatolıen (Mammalia, Rodentia). - An upper Pleistocene/lower Holocene rodent succession from alyasSWVeAnatolialMammaliayRodentia) 2. 2.2... une egn.n TıBA, T.; SaTo, M.; Hırano, T.; Kıra, 1.; SuGımuraA, M.; Suzuktı, Y.: An annual rhythm in reproductive activities and sexual maturation in male Japanese serows (Capricornis crispus). — Jahresrhythmische Veränderungen von Fortpflanzungstätigkeiten und Geschlechtsreife beim mannichenapanıschen Seraul(@apricomisenspus) no .eueeueeeeeenenen WINKING, H.; Durıc, BEATRICA; BULFIELD, G.: Robertsonian karyotype varıation in the European house mouse, Mus musculus. Survey of present knowledge and new observa- tions. — Karyotypvariation durch Robertsonsche Translokationschromosomen bei der euro- päischen Hausmaus, Mus musculus. Eine Übersicht über den derzeitigen Wissensstand und TEL Inormanomen. ee N er Re WIRTZ, P.; KAISER, PETRA: Sex differences and seasonal varıation in habıtat choice in a high density population of Waterbuck, Kobus ellipsiprymnus (Bovidae). — Geschlechtsunter- schiede und jahreszeitliche Variation in der Habitatwahl in einer Hochdichte-Population des DZisserbocksilXobusellipsiprymnus)) . u: 0 0 cn eeeunuonecen WOoLZ, IRMHILD: Ergebnisse automatischer Aktivitätsaufzeichnungen an Wochenstubenkolo- nıen der Bechsteinfledermaus (Myotis bechsteini). — Results of automatically monitoring Beelistemsbarsgaetivities 0... De a See N a. Zınsg, P. E.: Search calls of echolocating Nyctalus leisleri and Pipistrellus savıı (Mammalıa: Chiroptera) recorded in Switzerland. — Suchflugortungslaute von Nyctalus leisleri und arpısinellus saoı.(MammalıaChiroptera))inderSchweiz 2. 2m nn .uenn. 301 102 11 193 92 341 245 108 325 76 178 148 162 DIT, Wissenschaftliche Kurzmitteilungen DorT, MADELEINE van: Note on the skull size in the two sympatric Mouse Deer species, Tragulus javanicus (Osbek, 1765) and Tragulus napu (F. Cuvier, 1822). — Bemerkung über die Schädelgröße von zwei sympatrischen Hirschferkel-Arten, Tragulus javanıcus (Osbek, 1765) und Tragulus napu (F. Cuvier, 1822) ns BE an... 124 Draı, A. D. G.: A note on aquatic and aerial vision in Odontocetes. - Eine Bemerkung über das Sehen von Ödontoceti in Wasserund Luft... . 2. 2. 2 2 RE ae 55 FuLLer. T. K.: Bıknevicıus, A. R.; Kar, P. W.: Home range of an African wildcat, Fels silvestris (Schreber), near Elmenteita, Kenya. - Streifgebiet einer afrikanischen Wildkatze Felis silvestris (Schreber) nahe Elmenteita, Kenia .......... nr... 20m enuneen 380 "r Hart. L.: MOESKER, A.; VEDDER, L.; BREE, P. J. H. van: On the pupping period of Grey seals, Halichoerus grypus (Fabricius, 1791), reproducing on a shoal near the Island of Terschelling, the Netherlands. - Über die Geburtsperiode von Kegelrobben, Halichoerus erypus (Fabricius, 1791) auf einer Sandbank nahe der niederländischen Insel Terschelling.... 59 IBANEZ, C.: DELIBES, M.; CASTROVIEJO, J.; MARTIN, ROSALIA; BELTRAN, J. F.; MORENO, S.: An unusual record of Hooded seal (Cystophora cristata) in SW Spain. - Ein ungewöhnlicher Fund einer Klappmütze (Cystophora cristata)inSW-Spanien. ..... 2.2.2220 neon. 189 KIERDORF, U.; KIERDORF, H.: Weitgehende Rotation des 4. Prämolaren ım Unterkiefer eines Rothirsches (Cervus elaphus L.) und eines Rehbockes (Capreolus capreolusL.) ..... . . - SZ KURRE, ].; Fuchs, E.: Nachtaktivität von Spitzhörnchen (Tupaia belangeri). - Night activity of Tree shrews (Tupaiabelangen!) : ... . . : 2... ws. sun un de Pe 126 NeuHaus, W.: Zur Frage der Sehfähigkeit von Delphinapterus lencas in Wasser und in Luft. — On the question concerning the ability for vision of Delphinapterus lencas in water and inair. 57 WAEREBEEK, K. van; REYESs, J. C.: First record of the Pygmy killer whale, Feresa attenuata Gray, 1875 from Peru, with a summary of distribution ın the eastern Pacific. — Erster Nachweis eines Zwergschwertwales, Feresa attennata Gray, 1875, von Peru und eine Zusammenfassung über die Verbreitung im östlichen Pazifik .......... 2.2.2.0... 253 Bekanntmachungen SCHEN. 2 0. 2 02 a ee ee el ee east: 12 > 2 61, 382 Buchbesprechungen Seiten. PTR ATTENTAT TINTE 62, 128, 191, 256, 320, 384 This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con- tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugsweiser Verwertung, vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grundsätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheberrechtsgesetzes der 3undesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflich- tig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. : © 1988 Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstraße 12, D-2000 Hamburg 1; Lindenstraße 4447, D-1000 Berlin 61. Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., 2240 Heide/Holstein E ISSN 0044-3468 53 (1-6) 1-384 (1988) A, rg 53 (1), 1-64, Februar 1988 ISSN 0044-3468 C21274F =" EITSCHRIFT FÜR SAUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Organ der Deutschen Gesellschaft für Säugetierkunde König, A.; Rothe, H.; Sieß, Margaretha; Darms, K.; Gröger, Dagmar; Radespiel, Ute; Rock, J.: Reproductive reorganization in incomplete groups of the common marmoset (Callithrix jacchus) under laboratory conditions. — Reproduktive Reorganisation in unvollständigen Grupper@des Weißbüscheläffchens (Callithrix jacchus) unter Laborbedingungen Hansson, L.: Parent-offspring correlations for growth and reproduction in the vole Clethrionomus glareolus in relation to the Chitty Hypothesis. — Korrelationen von Wachstum und Reproduktion zwischen Eltern und Jungtieren bei der Rötelmaus Clethrionomys glareolus in Beziehung zur Chitty Hypothese Hickman, G. C.: The swimming ability of Ctenomys fulvus (Ctenomyidae) and Spalacopus cyanus (Octodontidae), with reference to swimming in other subterranean mammals. — Die Schwimmfähigkeit von Ctenomys fulvus (Ctenomyidae) und Spalacopus cyanus (Octodontidae) im Vergleich zu anderen subterranen Säugetieren 11 Dardaillon, Maryse: Wild boar social groupings and their seasonal changes in the Camargue, southern France. — Soziale Gruppenbildungen und ihre jahreszeitlichen Änderungen bei Wildschweinen in der Camargue, Südfrank- reich 22 Dubost, G.; Feer, F.: Variabilite comportementale & l’interieur du genre Cephalophus (Ruminantia, Bovidae), par "exemple de C. rufilatus Gray, 1846. — Verhaltensunterschiede innerhalb der Gattung Cephalophus (Ruminantia, —. N Bovidae) am Beispiel von C. rufilatus Gray, 1846 31 Furley, Ch. W.; Tichy, H.; Uerpmann, H.-P.: Systematics and chromosomes of the Indian gazelle, Gazella bennetti (Sykes, 1831). — Systematik und Chromosomen der Indischen Gazelle, Gazella bennetti (Sykes, 1831) 48 Wissenschaftliche Kurzmitteilungen Dral, A. D. G.: A note on aquatic and aerial vision in Odontocetes. —- Eine Bemerkung über das Sehen von Odontoceti in Wasser und Luft 55 Neuhaus, W.: Zur Frage der Sehfähigkeit von Delphinapterus leucas in Wasser und in Luft. - On the question concerning the ability for vision of Delphinapterus leucas in water and in air 97 t Hart, L.; Moesker, A.; Vedder, L.; Bree, P. J. H. van: On the pupping period of Grey Seals, Halichoerus grypus (Fabricius, 1791), reproducing on a shoal near the Island of Terschelling, the Netherlands. — Über die Geburtsperiode von Kegelrobben, Halichoerus grypus (Fabricius, 1791) auf einer Sandbank nahe der niederlän- dischen Insel Terschelling nn Bekanntmachungen U INN R Buchbesprechungen nr APR £ 1 1986 ÜBRARIES Verlag Paul Parey Hamburg und Berlin HERAUSGEBER/EDITORS P. J. H. van BREE, Amsterdam -— W. FIEDLER, Wien -— H. Frıck, München - W. HERRE, Kiel - H.-G. Krös, Berlin - H.-J. Kunn, Göttingen - E. Kurzer, Tübingen - B. Lanza, Florenz — J. NIETHAMMER, Bonn -— H. Reıcastein, Kiel - M. Röhrs, Hannover - D. STARcK, Frankfurt a. M. - F. Strauss, Bern — E. Thuenıus, Wien - P. VoceL, Lausanne SCHRIETLEILTUNG/EDITORTAT OFFICE H. SCHLIEMANN, Hamburg — D. Kruska, Kiel This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con- tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information Die Zeitschrift für Säugetierkunde veröffentlicht Originalarbeiten und wissenschaftliche Kurzmittei- lungen aus dem Gesamtgebiet der Säugetierkunde, Besprechungen der wichtigsten internationalen Literatur sowie die Bekanntmachungen der Deutschen Gesellschaft für Säugetierkunde. Verantwort- licher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Harald Schliemann. Zusätzlich erscheint einmal im Jahr ein Heft mit den Abstracts der Vorträge, die auf der jeweiligen Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde gehalten werden. Sie werden als Supplement dem betreffenden Jahrgang der Zeitschrift zugeordnet. Verantwortlich für ihren Inhalt sind ausschließlich die Autoren der Abstracts. Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z. Hd. Prof. Dr. Dieter Kruska, Institut für Haustierkunde, Biologie-Zentrum, Neue Universität, Olshausenstr. 40-60, D-2300 Kiel. Für die Publikation vorgesehene Manuskripte sollen gemäß den „Redaktionellen Richtlinien“ abgefaßt werden. In ihnen finden sich weitere Hinweise zur Annahme von Manu- skripten, Bedingungen für die Veröffentlichung und die Drucklegung, ferner Richtlinien für die Abfassung eines Abstracts und eine Korrekturzeichentabelle. Die Richtlinien sind auf Anfrage bei der Schriftleitung und dem Verlag erhältlich. Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der Korrekturfahnen erfolgen. Vorbehalt aller Rechte: Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei- ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grund- sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber- rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of the first page of an article in this journal indicates the copyright owner’s consent that copies of the article may be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copyıng beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective, or for resale. For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the GEg! Mit einer Beilage des Gustav Fischer Verlages, Stuttgart, sowie einer Beilage des Verlages Paul Parey Fortsetzung 3. Umschlagseite © 1988 Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstr. 12, D-2000 Hamburg 1; Lindenstr. 44-47, D-1000 Berlin 61. — Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., D-2240 Heide/Holst. Z. Säugetierkunde 53 (1988) 1-6 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Reproductive reorganization in incomplete groups ofthe common marmoset (Callıthrix jacchus) under laboratory conditions By A. Könıc, H. ROTHE, MARGARETHA SIESS, K. DARMS, DAGMAR GRÖGER, UTE RapesPiIeL and J. Rock Institute of Anthropology, University of Göttingen Receipt of Ms. 22. 12. 1986 Abstract Incomplete groups of Callithrix jacchus which exclusively consist of genetically related members show a rather considerable frequency of inbreeding. Therefore, the incest taboo seems to be realized at a very low level, eventually due to the laboratory environment. C. jacchus daughters are able to conceive in the presence of their mother. This result confirms to some extent the previous assumption upon a hierarchy dependent monogamy in the common marmoset. The reproductive reorganization of groups which have genetically unrelated members exclusively relies on these anımals. Introduction Most authors believe marmosets to be monogamous (e.g. ROTHE 1979). BıscHor (1985) describes their mating and reproductive organization as ‘arıstogamy’ since only the highest ranking group members (= parents) are allowed to interact sexually and to reproduce. Only from one species, 1.e. Sagnınns fuscicolhs, we have some information from field observations that tamarıns might be organized in cooperative polyandrous groups (TER- BORGH and GOLDIZEN 1985). Following the loss of one or both parent(s) monogamous groups are at first confronted with a temporary or even permanent loss of their ability to reproduce. The reproductive reorganization can be achieved by a subsequent breeding of the remaining parent with one of the adult offspring, however, undergoing by this all risks and disadvantages of inbreeding. On the other hand, reproduction can be continued by integration of a strange adult conspecific. Reproduction of the group ceases completely if neither inbreeding nor integration of a strange conspecific occurs. The same is true ıf the group dissolves by emigration of single famıly members or even subgroups. Up to now we do not have any detailed information on the strategy free ranging groups will follow when afflicted by the loss of one or both parents. Under laboratory conditions incomplete families as a rule cannot decide for a strategy of reproductive reorganization which includes gene flow by integration/immigration of a strange conspecific. That mode of regaining reproductive ability ıs only possible by interference of the investigator (ROTHE et al., in press). In this paper we confine to a description of inbreeding in the common marmoset and mechanisms of its avoidance. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5301-0001 $ 02.50/0 2 A. König, H. Rothe, Margaretha Sieß, K. Darms, Dagmar Gröger, Ute Radespiel, J. Rock Material and methods The data were taken from the diary of our marmoset colony and from observations during the daily routine work (e.g. observations on sexual behaviour). Informations on stability/instability of the respective families exclusively refer to those contexts which led to the expulsion or removal of one or even more group members (for details see ROTHE et al., in press). We analyzed 20 Callithrix jacchus groups which lived at least for another two months following the loss of their mother or at least two months after the last delivery of the mother in groups, whose «- male has died. The period of two months has been determined empirically, that is, we have made the observation that groups experienced dissolution shortly after the death of the «-female or the «-male (some days up to half a month). Therefore the chance to observe sexual behaviour has been very low. Furthermore all groups we describe in this paper had adult and fertile offspring. According to group composition we distinguish three categories: 1. groups which consisted of related members only (parents and their offspring) (n = 15). 2. groups with nonrelated members: a. by integration of hand- and/or foster mother reared infants/juveniles (n = 4); b. by integration of an adult male in an all-female group (mother and four daughters) (n = 1). Additional informations are given on two special groups which showed different basic parameters compared to the other families, but which revealed inbreeding as well. Results With the exception of the special groups sıx families (30 %) showed further reproduction (one to three litters). In two families we at least observed sexual behaviour. Groups with related members exclusively (n = 15) Reproduction continued in four groups. One of these became unstable after the loss of a parent whereas this was true for 75 % of those incomplete families which have ceased reproduction (Table 1). One group experienced unstable periods following the next delivery (Table 1). Table 1 refers to the group composition as well as to the dates of the breeding females’ conceptions. It is quite obvious that the unstable groups showed the longest break in reproduction. Group M is characterized by a specıal history (s. Table 1). The «-female has been sick for the last sıx months preceding her death. During that time she got rather regularly medical therapy. Already during her mother’s illness an adult daughter has been impregnated by her father. The resulting delivery occurred three months following the «-female’s death. ErpLE (1967) refers to a similar event in her C. jacchus colony. Groups with genetically unrelated members (n = 5) One group which had altogether three (2.1) integrated members (hand and foster mother reared infants) did neither show sexual nor reproductive behaviour. This group was remarkably unstable, however, we also recruted group members for pair formation. In two groups we observed copulations. Both families have been unstable periodically. After the death of the «-male copulations occurred between the «-female and one ot the genetically unrelated, meanwhile adult male group members. However, we could not detect any sexual intercourse between the mother and her adult sons. In another group a young female (11 months old) which had been integrated as infant into this famıly became pregnant after 45 days. This female was considerably younger than an adult daughter of the «-male, with whom the male did not interact sexually. In a third group we observed copulations between an adult male which has been integrated into the famıly some months ago and one of the oldest females shortly after the death of the a-female. About seven months after the death of the mother the female gave birth to triplets (see also RoTHE etal., in press). nn m nam nn mn pen nennen un m nie mann nme nam nern ame ne ° (Teaaoyur -UFATQISIUT 9Uo) PapnT>2ur yyırq/ssoT 194Je/91oJaq sÄep OLT "xoxdde o73 dn ATuo suotstndxa !ou g > Jur !ouw G 03 OT -_AnT !ow 0T 02 ST - ans !ou ST 03 GE APAISNJIXI sIaqwaw paJejı yym sdno1d) 'T aqrvL - peA !oW Sg < pe !UOTYNITOSSTP 93331duwoa - "SSTp 'dwos !(s)aaequau dno1Iß - WI !Teuwtrue-o MOU - %„ !SSOT 310oJaq Teutue-o - uU !5TewuoJ - J/oTeu - uw: (s)yyuou - ou EEE So Erin... 65 A1ISATITOp :poynAD91 J7J peÄ sAep ZI Ae2Je TI == == c6 ven SS... Zu = wur zu +3 3 We 0% x WU-O A OLE Et 69T OLT ou g’6 :"sstp *"dwoa er Orr urur = !u-o Jo ssot Ae47Je Äep auo AlsAaTtIsp I -— -_ Ze Ben = Sa _ - Ju wwgjjJu wugF v°’8 1 I) ke 66) ou TT :"sstp "duos - == — io >= u 8 1 = u = 33 u oc G'cc Zn dd Inn ou /T :"sstp "duosa >= 2 z !p33Jnı291 J Sauo pue uw ou g’FT 194Je - -- -- -- -- - on = ==. = — u = J33J u VII SHGE Sa 2) 070) BEN nee; ou s’TI :"sstp duo» ER {ei = !p3s3nı991 WI auo :3yßne>a sıayJuny oMmJ G Aep u0o -— -- -— SS = — € € -- .- - IT zu JuJF u 9°€E we = Bloß) ee) ou pz :"sstp "duo - —— 2 -- -- - -- - -- - wur = IT & {IT DIE 97 is) X Be irn Be Te Eee ee Eee ee nn wenn” aoy3uny UOTSTNndXa 2SATJ Je !W worJ uos u pe - m S— =—_ _— = a0 oo © — J Ju = wu ZASE 91 = ale) IN er me SCORE pRlauosniz3n zone ET Te WE a psSTp uw-o :3yßne> A9Juny UOTSITNdXS 4SATJ Je - -- -- e— == = g T u = = _ JJuJu wu En 1, = oe) fa gr A Er ou pz :'sstp "'duos - -- -- >= Zi = 6 [4 Sa 0 - = = yuur u WW 6° Zar D) au og ae aaa Des: Dee nn 1a37e SAep L PSTP 3-9 :g woıJ ıoy3yßnep J pe - — —— =... - -- - 7ı zZ - - - uF J ZU: 67 -- J/u-n Ta LOL 2 E u ou QT :"sstp "duwos :p3aTtTadxa RI om? 68 g Aep uo :DUTIUBTIF A94Je PspTATp sem dno1ß - = __ = _—_ = 8 7 co ce — _ uruI PPPTCFIUF 9°€E 99 ET /UOZg Er 7 STTaoH aoı dar m ea eu u ne en :3x93 99s ’dno1ß TetTaads wolJ p3apuasssap Ju T 0z 92 . __ = er So. © = = — I rU = Tat: ze = lie) ©@) p>3nI291 WO ee ee er ev ZELıng 9.5 I Ze u IBERHHrrRn . T :3yßnea Aoy9uny UOTSTNAX3 JSATJ Ye ZSIE /EN, IF Ju :UW-D JO SSOT 910J9q yyuow suUo AISATISP ISeI € SL L8T = S= = © 3 AuSSE — = zu 37 W„u} 955 G'’L6 Ze D) 9) PSATOSSTp dno16 ’umouyun AoyUJeJ T 92 sot == =— = -- . - -- - - - - „Ju „u ZIE 9E = Ko N c8 DLI = J-» Jo sSsoT 310J9q sAep 09 "xoadde uortydasuoa zZ 0% oo 8 00T 2 -- - -- — u uU JuUsJuU „u eg Es So Bloo) IM (ou) u2arq SSoT u £ sTlews3J SSOoT 19JJe 192Je sAep u sÄAep u (ou 2 en sAep sÄep SSOT SSOT dnoı 3ueu abe sÄep :y3ATq 19]2Je Aı33Je 910]J9q Jut Ant ans peÄ B: oTJeA Ri -Baad syıeuaıl :uotTJd3a>9uoD UOTSTNdX3 :W9 JO UOTSTNAXa sasset> xas/abe xas 3 J-o ssot dnoıß "uot310qge 193Je SAep 5 UOoT3de2uona :p>pnI>2urt 30u ’uor231oge SSOoT 193Je SsÄep 9 Doll aax> oTeu-n :dnoAß Terdads,., ou 77T :"sstp "duos !ps9nI2aaı WS ıayJo € !Jyßnea Aasyuny zZ !otgeysun Aısa eueaaq dnoa1ß ayF sÄep 002 "xoadde A193Je :!P33nA29A WI T SSOT A3JJe AÄTIAOUS(.9 :AASATISpP A92Je ITIT 9uweaag 2 !SSOoT 1934Je sSÄep L0OT AAISATTIP x :sAep 06T 'Xxoadde 193Je otTgea2sun oaweaaq dnoA !:SsoT A93Je SÄep TET AASATISP (y !,A9y3eJdajs ’pa3JeaßaJur J-0 Jo SSOT 3A10J39q ou 9 Ure :STEUTUE-D 3dOIX3 POATOSSTp sem dnolß AASATIOP 3SATJ A193Je !u Jo AayJoAaq u pe.z :4D dnoaß zZ oTqeI pue 4x33 995 dnoAß TeTaads puoass,|ı ° (TEAASJUT-YIATQAASIUT 9UO) PoSpnT>urT YIATq/ssot OT sAep OLT 'xoıdde SOsAnEATNONSUONSITNArS Mouse» Frurmllouschon OT - Anne louroTi oaNsT - anss lounsTroatgei- pe !ouU GE < pe !UOTINTOSSTP a39Tduoa - "sstp "duwoa !asquau dnoAß - WI :3UeJUI AO STTUSANT P9Aea1l A9UJOWU A194S0J A0o/pue -puey paxeaßsygur - + ZUOTIJ9eA9JUT TenxaSs Jo A9uUJIed AO TeEWTUeE-o MSU — y% !’ssoT SI10Ja4 Teurue-o - U !oTeusF - J !oTeu - w !(s)yyuow - ou 9 09 19 LT ccc 99 T9z I Zr a > + 334 zu *F IUEHSTETE RZ UueD (2d PN Se Te en Er SE a ee EEE Zee EN u ee ee — > ZZ = A SQ Fa re re zu w+u Fu u Va9 el -- 3-90 (ol ru = T > 2 S9 DIT cot oTz I Sn ZZSsE— Sur 33 zu 333 4 856 23 x U-0 (clD +J+U+U = ie > = 18 OTZ = Sa a 36. bb n = uur „+uUuU JF e’9 97 x Uu-n (vä 1 8t 29 == T € 602 IT GIVE: a ar = 33 I=J „u VASE 9 -- 3-9 (eNXUD v +ul = Ü 18JE Sr >= Zi On Zen Zr L ee - +HUrtF F = u ul SV, [bie -- 3-9 (zT (ou) uU3ATqQ SSOoT UYATQ SSOoT U sTeusJ SSOoT A93Je A92Je 19323Je 193Je sAep u sÄep u (ou 2 2230 sAkep sÄAep sÄep sÄep SSOT SSOT dnoı Jueu u ade sÄep :p3A1A3SqO :UYJATQ A92IFe 193Je 9»10J9q JuUT Ant ans peÄ Be orzel 33 -Baild UIJATq :uorTt3dae>uonD "aeyaq xX93Ss :ZUOTSTNnIdX3 :W9 JO UOTSTNdXa s3asset> xas/abde xas ade I-0 ssot dnoi1ß ‚sdno13 jernads pue sıaqwaw paJepps-uou yaMm sdno1d ‘7 a7gvL Reproductive reorganization in Callithrix jacchus 5 Special groups (n = 2) In one group the father has been so severely attacked by his son that he had to be removed from the family. He died some minutes later due to a severe shock. We do not know why the father has been attacked. Subsequently the son interacted sexually with his mother. This pair produced altogether sıx litters. In the second case a sexually experienced male (father of 13 litters) got for social companıon one of his adult daughters (mother of 7 litters). The female gave birth to male/female litter (see also FRENCH et al. 1984 for S. oedipus) which reproduced as well after removal of their parents (Table 1). Discussion The rather high frequency of inbreeding in incomplete groups cannot be interpreted as a mere accidential event. It is strıking how often the groups vıiolate the incest taboo, supposed such a phenomenon actually exists. The expulsion of the «-male in Gr. F (s. Table 2) (a sımilar situation was observed by SPICHIGER-CARLSSON, 1982) may indicate that the incest taboo ıs only weakly realized ın the common marmoset, and eventually influenced by the laboratory condition. No less important seems to us the sexual interactions of a female with her father and the onset of pregnancy whilst the mother was sick and had to be treated medically (see also ErpLE 1967). This observation confirms the findings of ABBorr (1984) and Evans and Hopges (1984) according to which the daughters may ovulate ın the presence of their mother. Besides this the mere physical presence of their mother (= highest ranking female in the family) does not seem to be sufficient to prevent sexual behaviour of the «-male with his daughter(s). The result of thıs ‘longterm study’ contrasts to the observations ın a “shortterm study’ of ANZENBERGER (1983), in which the presence of the mother and her offspring prevented sexual behaviour of the father and a strange female in a neighbouring cage, to which only the father and the strange female had access. However, we cannot exclude that the sickness of the «-female might have influenced the result, for example due to the eventuel loss of her a-status because of her physical inability. On the other hand ROTHEs (1974) hypothesis that marmosets are monogamous by status and not by emo- tional bond would be confirmed by that event. Most striking to us has been the fact that in groups with unrelated members the reproductive reorganization was not incestuous but was based on the integration of the genetically unrelated group members, even when they were younger than the offspring of the remaining parent. We had, however, to prove, whether there has possibly existed a dominance-subordination relationship between the integrated and the family-born group members. In this case the avoidance of inbreeding could only be interpreted as a secondary phenomenon. But if incest avoidance must be regarded as a primary event, then the cognitive capacity of the common marmoset must be highly valued, especially since no group odor could be made responsible for that result (EPPLE, pers. comm.), except it would be genetically determined by a single Mendelian gene locus. To what extent young females experience an accelerated sexual maturation when becoming «-female cannot be answered at the moment (see also TARDIF 1984). We do not know whether infant transfer and/or infant-emigration/-immigration can be regarded as a regular event in the life of a marmoset group in order to offer a proper strategy for the reproductive reorganization of an uncomplete family. However, Dawson (1976) observed a rather frequent migration of juvenile Saguinus oedipus geoffroyi between . neighbouring groups. At least in this tamarın species infant transfer seems to be a regular behaviour in the natural habitat. 6 A. König, H. Rothe, Margaretha Sieß, K. Darms, Dagmar Gröger, Ute Radespiel, J. Rock Zusammenfassung Reproduktive Reorganisation in unvollständigen Gruppen des Weißbüscheläffchens (Callithrix jacchus) unter Laborbedingungen Die vorliegende Arbeit beschreibt Inzucht und deren Vermeidung in unvollständigen Callthrix jacchus-Gruppen. Untersucht wurden 20 Gruppen, die in drei Kategorien gegliedert wurden: l. Gruppen, die ausschließlich verwandte Mitglieder enthalten; 2. Gruppen mit genetisch fremden Tieren a. entweder durch Integration infantiler/juveniler Tiere oder b. durch Integration eines adulten fremden Männchens in eine Weibchen-Gruppe (Mutter und vier Töchter). Insgesamt sechs Gruppen setzten die Reproduktion nach Verlust eines Elters fort. Die relativ große Anzahl von Familien (4 von 15 Gruppen), in denen nach Verlust des Elters Inzucht auftrat, kann nicht mehr als ein rein zufälliges Ereignis gewertet werden. Sehr auffällig ist die Einbeziehung der genetisch nicht verwandten Tiere in die reproduktive Reorganisation, in solchen Gruppen (n = 5), in die genetisch fremde Tiere integriert worden sind. References ABBoTT, D. H. (1984): Behavioral and physiological suppression of fertility in subordinate marmoset monkeys. Amer. J. Primatol. 6, 169-186. ÄNZENBERGER, A. A. (1983): Bindungsmechanismen in Familiengruppen von Weißbüscheläffchen (Callithrix jacchus). Inaugural Diss., Zürich. BıscHor, N. (1985): Das Rätsel Odipus. Die biologischen Wurzeln des Urkonfliktes von Intimität und Autonomie. Piper: München. Dawson, G. (1976): Behavioral ecology of the Panamanıan tamarın, Saguinus oedipus. University Microtilms Int., p. 1-164, Ann Arbor, Michigan. EpPL£, G. (1967): Vergleichende Untersuchungen über Sexual- und Sozialverhalten der Krallenaffen (Hapalidae). Folia primatol. 7, 37-65. Evans, $.; Hopges, J. K. (1984): Reproductive status of adult daughters in family groups of common marmosets (Calhıthrix jacchus). Folia primatol. 42, 127-133. FRENCH, ]. A.; ABBOTT, D. H.; Snowpon, C. T. (1984): The effect of social environment on estrogen excretion, scent marking, and sociosexual behavior ın tamarıns (Saguinus oedipus). Amer. ]. Primatol. 6, 155-167. ROTHE, H. (1975): Some aspects of sexuality and reproduction in groups of captive marmosets (Callıthrix jacchus). Z. Tierpsychol. 37, 255-173. — (1979): Das Ethogramm von Callıthrix jacchus Erxleben, 1777 (Primates, Ceboidea, Callitrichi- dae). Eine morphaktische Analyse des Verhaltens mit besonderer Berücksichtigung des sozialen Umfeldes. Habil.schrift, Göttingen. ROTHE, H.; Köntc, A.; Darms, K.; Sıess, M. (in press): Analysis of litter size in a colony of the common marmoset (Callıthrix jacchus). ROTHE, H.; Köntc, A.; Sıess, M.; RADESPIEL, U.; Darms, K. (1987): Integration adulter Callıthrıx jacchus-Männchen in Rumpfgruppen. (in press). SPICHIGER-CARLSSON, P. (1982): Beziehungen unter Familienmitgliedern bei Weißbüscheläffchen. Lizentiatsarbeit, Universität Zürich. TAarDıF, $. D. (1984): Social influences on sexual maturation of female Saguinus oedipus oedipus. Amer. ]. Primatol. 6, 199-209. TERBORGH, J.; GOLDIZEN, A. W. (1985): On the mating system of the cooperatively breeding saddle- back tamarın (Saguinus fuscicollis). Behav. Ecol. Sociobiol. 16, 293-299. Authors’ address: ANDREAS KÖNIG, HARTMUT ROTHE, MARGARETHA SIESS, KURT DARMS, DAGMAR GRÖGER, UTE RADESPIEL, JEns Rock, Institut für Anthropologie, Universität Göttingen, Bürgerstraße 50, D-3400 Göttingen Z. Säugetierkunde 53 (1988) 7-10 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Parent-offspring correlations for growth and reproduction in the vole Clethrionomus glareolus in relation to the Chitty Hypothesis By L. Hansson Department of Wildhfe Ecology, Swedish University of Agricultural Sciences, Uppsala Receipt of Ms. 13. 11. 1986 Abstract Studied in laboratory correlations between parents and offspring for weights and reproductive parameters in voles Clethrionomys glareolus trom a non-cyclic population. Weight correlations were significantly positive in males and negative in females. Reproductive parameters correlated negatively. The latter correlations could be analysed further and be interpreted as due to a negative maternal effect. These findings contradict the Chitty Hypothesis of population regulation but may explain cyclic variations in weight and reproduction if external factors drive the cycles. Introduction Microtine cycles are characterized by small anımals with high reproductive output during the increasing phase and large, slowly-reproducing anımals at peaks. However, small adult animals wıth low reproduction occur during the declining phase (Kress and Myers 1974; TAıtT and Kress 1985). The Chitty Hypothesis (CHıttyY 1967, 1970, cf. also Kress 1978), predicting regular genetic changes with cyclic fluctuations, has had a profound impact in explaining such morphological and reproductive varıations. It states that timid, slowly-growing but high- reproducing anımals are favoured in sparse and ı increasing pupulations whereas aggressive, large but slowly-reproducing anımals are favoured in dense or peak populations. This mashinery should drive the cycles. Body growth and reproductive rates should thus be mainly genetically determined in microtine populations; body size and reproductive output should be closely correlated between parents and offspring. Animals caught in varıous population phases and brought to a laboratory should retain such correlations ıf there is low or random (density-ındependent) selection in the laboratory. These predic- tions were examined on the bank vole Clethrionomys glareolus which appears in both cyclic and non-cyclic populations in Sweden (Hansson and HENTTONEN 1985a). Methods The study was performed on anımals which were bred for other reasons (Hansson and HENTTONEN 1985b; Hansson 1986) under laboratory conditions. The founding anımals were taken from a clearly non-cyclic population in south Sweden (Hansson and HENTTONEN 1985a). Thus, it was possible to evaluate parent-offspring relations with regard to the pattern of population dynamics. The data were examined according to the rules of quantitative genetics but the relationships were mainly expressed as correlations (cf. MıLLAR 1983) since certain prerequisites for heritability analysis may not have been present. Bank voles were caught in live traps at Revinge (56°N) in 1980-83. The anımals were kept as monogamous pairs after capture so the low mortality (19 and 8 % per year for wild-caught and m U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5301-0007 $ 02.50/0 8 L. Hansson laboratory-born animals respectively) was not density-dependent. The voles were caught in early autumn as young anımals and kept on constant food (laboratory mouse pellets) and at a constant temperature (20°C) for one year or until reproduction ended. Young of these animals, born in late summer-early autumn, were kept as monogamous pairs from an age of four months under the same laboratory conditions. Males were weighed every second week and the maximum weights recorded for each individual was used for the computations. In females, pregnancies caused exceptional weights so instead the weight one month before the first parturition was used. The length of pregnancy is ca 20 days in this species. Times of first parturition, litter sizes and number of litters as well as the total number of weaned young were recorded for all field-caught females (P) and their female offspring (F,). Correlations were examined between weights and reproductive parameters of field-caught animals and the corresponding means of their laboratory-born offspring (FALCONER 1981). Results and discussion Males showed a strong positive correlation in weights between parents and offspring (Table 1). Females showed a negative correlation, on the border of significance, for the same relationship. The number of litters, number of weaned young, mean litter size and start of reproduction were all consistently Table 1. Correlations between maximum body but non-significantly negatively related be- weight (see text for estimation) of Pand FR tween female parents and offspring Clethrionomys glareolus (Table 2). There were thus clearly positive parent- otfspring relationships ın the body weights Me on on of the males. The mean weight (X #SD) ot Bee Soon parent (P) males was 24.3 + 2.4 g and of otfspring (Fı) males 23.8 # 3.0 g so there was no significant change due to laboratory breeding. These relations have to be interpreted as genetic as no sources of error would cause a change in that direction for males (cf. below for females). The regression coefficients (b+SE) were 1.02 + 0.24. According to FALCONER (1981), this indicates a Sex IE Pe Table 2. Correlations between reproductive parameters (see text for estimation) of P and F} Clethrionomys glareolus Reproductive parameter Number of litters Weaned young Litter sıze Start of reproduction very high level of heritability. The generally negative relations in both body weights and reproductive parameters between female parents and offspring have to be interpreted as “a negative maternal effect” in analogy with LEAmYv (1981) and Mırrar (1983). Both these authors got consistent but low and often non-significant negative correlations in these parameters for both Peromyscus leucopus and P. maniculatus. They suggested that these negative correlations were due to well-fed females giving birth to large litters, where however each young had a low body weight. The female young were supposed to retain a comparatively low body weight as adults, to reproduce late and to produce small litters. The applicability of this interpretation was examined as regards litter sizes and weight at weanıng (Table 3). Litter sızes were significantly larger in the wild-caught (P) than in the laboratory-bred (F,) females while female young at weaning (20 days) were heavier, although not significantly so, in the litters of F, females. However, a significantly lower Parent-offspring correlations for growth and reproduction in the vole 9 Table 3. Litter size and female weaning weight in litters produced by P and F, Clethrionomys glareolus females, weighed before breeding started Weight before breeding Litter size Weights of female young at weaning Mothers N x SD Significancee N x SD Significance N x SD _ Significance 1D SD Zoll 141 s h NE P<0.001 190,05 NS IE 3 85 - 1.4 710: O5 weight had appeared at maturity in the F, females from P litters. Thus a negative effect due to alternating large and small litters was supported also by the present data. It is obviously due to limited body resources (energy or nutrients) and the mobilizable amounts of these resources are probably related to body size. The correlations in Table 2 are not significant. However, the interpretation of them as maternal effects on reproduction is supported both by similar findings in two American small rodent species and by significant differences in a derived relationship between litter sızes and female weights. Thus, although the reproductive effects are not as evident as the correlations in body weight both may be used for evaluating ıdeas relating to weight and reproduction in free-living populations. One of the correlates of the proposed mechanisms in the Chitty Hypothesis for cyclic vole populations, i.e. a strong heritability of male body weights, was thus observed also in a non-cyclic vole population, thereby indicating a lack of importance of this factor in population dynamics. The negative maternal effect may be important in the demography of eyclie populations if the cyclicity is caused by other factors. In the present study it was observed under the same nutritional conditions for parents and offspring. However, the body growth improved in animals taken into the laboratory from the field and given surplus food (Hansson 1985, unpubl.). In an increasing vole population there is more food per individual than ın a peak population. Thus, the negative maternal effect should be much more obvious under field conditions in peak/decline populations. It may at least partly explain why increase-early peak anımals reproduce early and have larger litters (Hansson and HENTTONEN 1985b) and why late peak-decline anımals show low body weight and late start of breeding, as evident for C. glareolus in, e.g., Hansson (1984). Summarizing, selective effects on individual characteristics related to body size do not cause vole cyclicity according to the Chitty Hypothesis while reproductive patterns in vole cycles, also appearing in the Chitty Hypothesis, may be explained by alternating positive and negative maternal effects. Acknowledgements I am grateful to BÖRJE Larsson for genetical advice. The work was supported by the Swedish Natural Science Research Council. Zusammenfassung Korrelationen von Wachstum und Reproduktion zwischen Eltern und Jungtieren bei der Rötelmaus Clethrionomys glareolus in Beziehung zur Chitty Hypothese An nicht zyklischen Populationenen von Clethrionomys glareolus wurden Korrelationen von Wachs- tum und Reproduktion zwischen Eltern und Jungtieren untersucht. Die Wechselbeziehungen in der Körpergröße waren positiv bei männlichen und negatıv bei weiblichen Individuen. Die Wechselbezie- hungen in bestimmten Fortpflanzungsparametern waren negativ und wurden weiter analysiert. Sie könnten durch einen negativen maternalen Effekt bedingt sein. Diese Befunde widersprechen der Chitty Hypothese zur Populationsregulierung, könnten aber die zyklischen Variationen von Gewicht und Reproduktion erklären, wenn äußere Faktoren die Zyklen beeinflussen. 10 L. Hansson Literature CHiıtty, D. (1967): The natural selection of self-regulatory behaviour in anımal populations. Proc. Ecol. Soc. Australia 2, 51-78. — (1970): Variation and population density. Symp. Zool. Soc. London 26, 327-334. FALCONER, D. $. (1981): Introduction to quantitative genetics. 2nd ed. London: Longmans. Hansson, L. (1984): Composition of cyclic and non-cyclic vole populations: On the causes of variation in individual quality among Clethrionomys glareolus in Sweden. Oecologia 63, 199-206. — (1985): Geographic differences in bank voles Clethrionomys glareolus in relation to ecogeographi- cal rules and possible demographic and nutritive strategies. Ann. Zool. Fennici 22, 319-328. — (1986): Geographic differences in the sociability of voles in relation to cyclicity. Anım. Behav. 34, 1215-1221. Hansson, L.; HENTTONENn, H. (1985a): Gradients in density variations of small rodents: The importance of latitude and snow cover. Oecologia 67, 394402. Hansson, L.; HENTTONEn, H. (1985b): Regional differences in cyclicity and reproduction in Clethrionomys species: Are they related? Ann. Zool. Fennici 22, 277-288. Kress, C. J. (1978): A review of the Chitty Hypothesis of population regulation. Can. J. Zool. 56, 2463-2480. Kress, C. J.; Myers, J. H. (1974): Population cycles in small mammals. Adv. Ecol. Res. 8, 267-399. Leamy, L. (1981): The effect of litter size on fertility in Peromyscus leucopus. J. Mammalogy 62, 692-697. MıLLAR, J. S. (1983). Negative maternal effects in Peromyscus maniculatus. J. Mammalogy 64, 540-543. TaıtT, M. J.; Kress, C. ]J. (1985): Population dynamics and cycles. In: Biology of New World Microtus Ed. by R. H. Tamarın. Spec. Publ. Amer. Soc. Mamm. 8, 567-620. Author’s address: LENNART Hansson, Department of Wildlife Ecology, Swedish University of Agricultural Sciences, $-750 07 Uppsala, Sweden Z. Säugetierkunde 53 (1988) 11-21 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 The swimming ability of Ctenomys fulvus (Ctenomyidae) and Spalacopus cyanus (Octodontidae), with reference to swimming in other subterranean mammals By G. C. HıcKMAN Department of Zoology, University of Natal, Pietermaritzburg Receipt of Ms. 21. 5. 1986 Abstract Investigated the capacity for swimming for the South American rodents, Ctenomys fulvus and Spalacopus cyanus. Like most non-cricetid and non-murid rodent moles which have been tested, both Ctenomys and Spalacopus were not strong swimmers (mean swim time ISSUE on IL SS I EUR SSR NN se IE EN N x AS. Nu" ya vr z= Da ze Sera Ir —; 2 a 4: —— >> Ge Se en “Is —— Pe Fig. 1. Various views of Ctenomys fulvus swimming in an aquarıum at 23 °C (drawn from photo- graphs). Above: Side-view: note the horizontal body posture with snowshoe-like hindfeet; both fore and hindfeet with coordinated movements of the robust, tactile, tail (seen here partially lifted above the water) are also important in the transport of excavated soil. Middle: Three-quarter front view: the large foreclaws did not obstruct stroking, and the head with typically enlarged rodent incisors is maintained well above water level. Below: Top-view: only slight lateral movements of the hindquar- ters were noted; the taıl was not used for sculling 14 G. ©. Hickman \ I & x N, UST UN SI ATI NUN RER DIES RS DI? I In! AN OR NII = z EEE UN 5% RT - FRI Ss K ir III aNS NS Fig. 2. Side view of Spalacopus cyanus during an aquarium trial at 23 °C (drawn from a photograph); note that the hindquarters remained low in the water, despite hindfeet positioning close to the body and strong stroking provided by all four limbs propulsion); the dog-paddle switched to synchronous hindleg kicking when swimming rapidly in linear paths, although the forelimbs continued to stroke alternately. Spalacopus paddled strongly, so that ıt is surprising that the hindquarters remained low in the water during swimming. Swimming speed reached a maximum of 30 cm/s, wıth turns being executed with little apparent difficulty. No attempts at digging were made by anımals when returned to the cages; anımals either rested quietly or groomed themselves. Discussion Adaptations of marine mammals to swimming has been well documented for some time (Hower 1930), while the swimming of most terrestrial mammals has remained almost totally unrecorded (Dass and Wınpsor 1972). Except for some early observations on talpids (FisHER 1885; MERRIAM 1884; FooT 1941; REED and RHuıney 1943), the swimming ability of all subterranean mammals was unknown until recently (see Table), despite the Table. Classification of subterranean mammals tested for swimming ability, habitats sampled, projected width (m x min) of traversable water barriers, and sources of data Order Family Gernera tested Capture site Projected References distance (m) Insectivora Talpidae Condylura Lake shore 1440 Hiıckman 1984b Parascalops Lake shore 1080 Scalopus Meadow 150 | Chrysochloridae Amblysomus Grassland 360 Hiıckman 1986 | Eremitalpa Desert = | Chrysospalax Forest 8 | Rodentia Geomyidae Thomomys Mountains 84 Hiıckman 1977 Geomys Plains 26 | Pappogeomys Plains 18 | Bathyergidae Heterocephalus Semi-desert 54 Hiıckman 1983b | Cryptomys Grassland 35 _Hıckman 1978 | Rhizomyidae Tachyoryctes Grass, few trees 40 HıckMan 1983a Octodontidae Spalacopus Mountains 38 Present study | Ctenomyidae _Ctenomys Semi-desert 9 Present study | Spalacıdae Spalax | Mountains, 1 HıcKMAaN etal. desert 1983 The swimming abılıty of Ctenomys fulvus and Spalacopus cyanus 15 fact that moles and mole-rats are generally very poor climbers (HıckMmAn 1982) and are thereby committed to swimming should water rise to sufficient levels. Flat areas, and highways (HueEy 1941), and rıvers (KENNERLY 1963; SMITH and PATToN 1980) are moist areas which form corridors for dispersal, and are habitats particularly vulnerable to flooding. Normally non-vagile due to the high energy costs associated with excavating burrows (VLEcK 1979), subterranean mammals may thus be induced to disperse from established areas and facilitate gene flow as anımals “hurdle” edaphic barriers while swimming. Moreover, colonization is facilitated as re-establishment of burrows is not excessively difficult in moist and friable soils upon reaching the shoreline. The above scenario must however, meet one very important basic requirement: the anımals must be able to swim. Rafting does not appear to be of much consequence, since subterranean mammals made no attempt at clinging to or climbing on buoyant debris provided during some of the testing. The swimming ability of Ctenomys fulvus, representing one of the last major eutherian families of subterranean mammals to be tested (Table), reinforces the general pattern which has emerged from previous studies: insectivore moles are, as a general rule, better swimmers than rodent-moles (Fig. 3). For those species tested, the mean swimming time for insectivore moles was 20 min to only three minutes for rodent moles and swimming rate 21 m/min to 12 m/min in favour of insectivore moles, resulting in a projected (m X min) overall advantage in dispersal distance of 608 m to only 34 m to the advantage of insectivore moles. The above figures are likely minimal, as favourable currents and larger sample sizes over many years would undoubtedly reveal greater dispersal powers. None- ee 40 ” Sri = wi = = [®} p4 = = 20 4 7) 36 64 10 15 Ss) RYZ BTH GEO OCT CTN SPA TB 3 a ee ee a re INSECTIVORA RODENTIA Fig. 3. Mean swimming times recorded for several species belonging to the major groups of eutherian subterranean mammals under laboratory conditions are indicated by bars; lines atop bars indicate the upper ranges during trials; the number of trials are given at the top of each line. TLP = Talpidae; CHR = Chrysochloridae; RYZ = Rhizomyidae; BTH = Bathyergidae; GEO = Geomyidae; OCT = Octodontidae; CTN = Ctenomyidae; SPA = Spalacidae. References for data and species for each column are listed in the Table 16 G. C. Hickman theless, laboratory testing has indicated an ability to traverse at least small waterways which may form an effective barrier to even surface rodents (SAvAGE 1973). Insectivore vs Rodent moles Size and pelage appear to be primarily responsible for the dichotomy in swimming performance between insectivore and rodent moles. The smaller size and weight of insectivore moles (a mean of 100 g to 146 g for rodent moles tested, see Fig. 4) results in a 400-1] ®(I)Chrysospalax 300 ®@(R)Pappogeomys ® (R)Ctenomys 200 ®(R)Geomys MASS (GM) ® (R)Tachyoryctes © (R)Spalax ®@ (R)Spalacopus 100 o( R)Thomomys © (1)Amblysomus © (R)Cryptomys e(1)Scalopus ® (R)Heterocephalus e(l\Parascalops @& (1)Condylura e(!)Eremitalpa 0 10 20 30 20 >3 MEAN SWIMMING TIME (MIN) Fig. 4. Mean body mass of species plotted against mean swim times of insectivore (I) and rodent (R) moles small body mass to surface ratio, with air trapped in the pelage exerting a greater influence on the buoyancy of swimming animals; the fur of talpids (the most successful subterranean mammal swimmers tested thus far) is renowned for thickness and insulative properties which not only assists floating, but protects talpıds (which are small in size) from rapıd temperature loss to the water. The uncharacteristic poor performance by an insectivore mole, the giant mole Chrysospalax, was largely attrıbutable to the inability of these large anımals to float (Hıckman 1986), while the smallest of the rodent-moles tested (Heteroce- phalus) did not perform well due to the lack of fur and subsequent lack of floating ability (HıckMman 1983b). None of the rodent-moles were able to float effectively due to the large skull and massive head musculature important to digging and gnawing vegetation; insecti- vore moles do not dig with the head and have lighter skulls with only moderately developed jaw musculature for crushing insects and severing earthworms. Moreover, talpıds have an elongate snorkle-like snout which enables the nostrils to be held high above the water surface (HıckmAan 1984b). Other non-subterranean rodent species such as the fossorial muskrat Ondatra zibethica are good swimmers despite a large size (WILBER 1958, 1963), so that factors other than size and pelage must also contribute to swimming success for mammals in general. Anatomical features other than size and pelage appear of less importance to the swımming ability of subterranean mammals. Although insectivore moles (the best The swimming abılıty of Ctenomys fulvus and Spalacopus cyanus 17 swimmers) are more characteristically blind than rodent moles, the only rodent-mole with eyes completely covered with skin (Spalax) was one of the worst of all mammal swimmers; all subterranean mammals, whether effective swimmers or not, maintained equilibrium when swimming. The more prominent mystacial vibrissae of rodent moles, important to other species of rodents during swimming (AHr 1982), appeared only minimally to assist with balance. The “comb” fringing the lateral edge of the foot of Ctenomys fulvus did not play a significant role for propulsion or balance. Although the best swimmers (the talpids) were tailed, the next best swimming group (the chrysochlorids) lack any form of external tail; and although the worst rodent-mole swimmer (Spalax) lacks an external tail, the best rodent-mole swimmers (Cryptomys) have only a short, stubby tail. The tail is of some assistance as a probe and for maintaining balance in some moletypes not only during digging (HıckMmAn 1984c) but also during swimming (Best and HART 1976; HıckmAan 1979); the taıl was utilized for propulsion when swimming only by Condylura, and then only during diving (Condylura was the only species tested which attempted diving). The development of enlarged foreclaws for digging had little effect on swimming (geomyids vs bathyergids, for example), but the partial webbing of the digits common to many subterranean mammals (responsible for the misnomer Scalopus aquaticus, a non- aquatic species), normally utilized when pushing soil, appeared to contribute to effective stroking I in the water (HanAawALT 1922). Other preadaptations to swimming common to both insectivore and rodent moles include a streamlined body contour, strong nasal sphincters, ability to withstand low oxygen levels, and powerful limb musculature. The fact that Ctenomys differs from most subterranean mammals by transporting excavated soil to the surface by kicking with the back legs rather than turning around and pushing the soil with the anterior portion of the body (Hıckman 1985) had no discernable effect on either the manner or ability to swım. In fact, most behavioural differences had little influence on swimming ability. A few insectivore and rodent moles remained motionless and floated instead of trying to swim, so that the fur wetted less quickly due to less water turbulence; however, immobile anımals would stand little chance of reaching the safety of the shore under natural conditions. Spalax was one species where a few individuals were able to improve on swimming performance by utilization of a characteris- tic “head lift” (Hrckman et al. 1983). Predicting swimming ability With convergent anatomical adaptations and little behavioral differentiation in swimming behaviour, how predictable ıs a species performance in water? T'wo general trends emerge from studies on the swimming ability of subterranean mammals: 1. insectivore moles swim better than rodent-moles; 2. smaller species of mole types swim longer than larger mole types. There are, of course, exceptions to the trends; the large size of Chrysospalax for an insectivore mole is the exception to the first trend which substantiates the second trend. Yet, one of the few species of any mammal mole-type to adapt to an aquatic existence, the talpıd Desmana (250-400 g), has an adaptation of many aquatic mammals: large size (relatively speaking) for low surface to body mass ratio which conserves heat and offers some degree of protection from predatory amphibians (Pıne 1975), fish, or birds. On the other hand, the Pyrennean desman (Galemys, 50-80 g) retains the small size of most subterranean mammals, as has an aquatic insectivore shrew, Neomys (10-20 g). Surprising- ly, although talpıds include the aquatic Desmana and semi-aquatic Condylura, the other insectivore mole-group, the chrysochlorids, have not radiated into even the semi-aquatic habıt (Hickman 1986), and the only reference found for any group of rodent-mole was for the ctenomyid Ctenomys lewisi (WALKER 1975 p. 1047) which, “... tunnels in stream banks and may be semi-aquatic”. There is little to suggest in the present study on C. fulvus 18 G. C. Hickman or any other rodent-mole that any species of rodent-mole regularly visits water and swims, despite the promise of plentiful food resources, a means of escaping predators, and extensive energy savings (VLECK 1979) by circumventing the need for excavating extensive tunnel systems. No mole-type has widely colonized the marine littoral environment (McCurrery 1967) although some species (Bathyergus) may be found in dunes. Since the subterranean environment is stable, reliable, and safe (most moles are K-selected), adaptive pressures would have to be very strong to promote abandonment of the sub-terrestrial niche and the risks involved with numerous novel situations. Perhaps habitat differences between varıous mole types may give some indication of swimming ability. Subterranean mammals establish tunnel systems in a wide variety of habitats (Table), with insectivores more characteristically inhabiting moister areas where invertebrates are more numerous. Rodent-moles are also abundant in moist areas where there ıs an abundance of plants to eat, but are also found in drier areas less characteristic of insectivore moles where tubers, bulbs, and other geophytes are available. Insectivores from moist areas are much better swimmers, even swimming through water-filled tunnels (Moore 1939). Rodent-moles (Spalax) from mountainous areas with seasonal flooding are better swimmers than con-specifics from desert habitat (Hıckman et al. 1983). HickMAN (1977) also suggests that regular spring-tlooding in mountainous areas would select for swimming adaptations in geomyids (Thomomys were the best geomyid swimmers tested); however, BAINTER and HarT (1979) cite good drainage in mountainous areas as not selecting for swimming ability. In any event, much water is seasonally available in mountainous areas, and a potential intluence on swimming ability. Habitat may affect body size, amount of fat deposits, development of musculature, oiliness of the pelage, and many other physical characteristics which might indirectly atfect buoyancy and swimming; however, no systematic studies have been conducted to determine the effect of these factors, and predicting swimming ability remains tenuous at best. Moreover, there is a need for wider experimentation with a wider range of mammalian types before further generalities can be formulated and the significance of the results appreciated. In comparison with some rodent species (ESHER et al. 1978), insectivore moles compare quite favourably, yet other rodents such as Microtus swim at least as well as insectivorous moles (CARTER and MERRITT 1981; FISHER 1961), a characteristic which has enabled colonization of coastal islands. The value of studies on swimming Asıde from practical value (evaluating flooding as a control measure for mole-types in agrıcultural or horticultural areas, as a test for the effectiveness of drugs (SHAPIRO et al. 1970), or as a water barrier for zoological gardens), there is heuristic value in having a comprehensive ethogram for mammals; many numerous and widely distributed anımals (Ctenomys, for example, which extend almost the entire length of South America) have remained totally untested for swimming ability. Swimming ability, not as an end ın itself but in conjunction with other factors, can be an important tool to zoogeographic analysıs (SMITH and PaTron 1980; WıLkıns 1985), enabling more knowledgeable interpretation of such behaviours such as position of the nest and direction of tunnel extension (MILLER 1957; WırLıams 1976). Thus far, studies on the swimming ability of subterranean mammals have involved the importance of various morphological, behavioural and abiotic factors (HıckMmAn 1978, 1982, 1983b) to swimming success; phylogenetic affinities by use of behaviour as a taxonomic tool (HickMan 1977); the question of an aquatic vs terrestrial origin for talpids (Hıckman 1984a); examination of the question why subterranean mammals have not more widely radiated into aquatic niches (HıckmAn 1986); the ecology of seasonal flooding (Hıckman 1983a); and the importance of water as a barrier to dispersal and subsequent speciation (KEnNnERLY 1963; HIcKMman et al. 1983). The swimming ability of Ctenomys fulvus and Spalacopus cyanus 19 Concerning mammals as a group, areas of study such as the ontogeny of swimming ability remain largely unexplored (Kıng 1961; ScHAPIRO et al. 1970; DacG and WINDSOR 1972), while further studies are needed to compare swimming performances within genera (SCHMIDLEY and PACKARD 1967; HAFNER and HAFNER 1975), within (Evans et al. 1978) or between (HıckMman and MACHINE 1987) families, and even between orders (STARRETT and FisHEr 1970). Much further basic experimentation and understanding in terms of events which occur only rarely or remain for the most part unobserved is needed for a full appreciation and evaluation of free water as a significant feature in dispersal, adaptive radiation, and general biology of subterranean mammals, or indeed, of mammals in general. Acknowledgements In Chile, Dr. Marıo RosEnMAnN and Sr. RIGOBERTO SoLis of the Universidad de Chile were most helpful with equipment, advice, and translations. Lu1s CoNTRERAS kindly procured live tuco-tucos and coruros and arranged for safe shipment of the anımals to South Africa; the C.S.I.R. made the study possible with a post-doctoral bursary and travel allowance, in conjunction with the University of Natal which provided a leave furlough with travel grant. SusiE, my wife, wore the cap of secretary, mother, adviser, and not least of all, cheerleader, throughout the study. GORDON MACLEAN undertook the translation of English to German, and JuLıE Cook kindly transformed pages of scıbble to legible text. The kind assistance of everyone during this and associated studies is gratefully remembered. Zusammenfassung Die Schwimmfähigkeit von Ctenomys fulvus (Ctenomyidae) und Spalacopus cyanus (Octodontidae) im Vergleich zu anderen subterranen Sängetieren Die Schwimmfähigkeit von Ctenomys fulvus (Ctenomyidae) und Spalacopus cyanus (Octodontidae) wurde untersucht. Wie die meisten anderen bisher untersuchten, wühlenden Nager sind diese Arten mit durchschnittlich unter 2 min Schwimmzeit wenig ausdauernde Schwimmer. Hinderlich bei Spalacopus war offensichtlich sein schnell durchnässendes Fell. Ein Vergleich der Schwimmfähigkeit bei wühlenden Nagern und Insektenfressern ergab: Wüh- lende Insektenfresser (untersucht wurden Talpidae und Chrysochloridae) sind im Durchschnitt bessere Schwimmer als wühlende Nager. Innerhalb der Gruppen wühlender Nager hat sich keine Art an das Wasserleben angepaßt, wie das die Desmaninae innerhalb der Talpidae getan haben. Literature Ant, A. S. (1982): Evidence of use of vibrissae in swimming Sigmodon fulviventer. Anim. Behav. 30, 1203-1206. BAINTER, S. L.; HART, E. B. (1979): Swimming ability of the northern pocket gopher. Proc. N.D. Acad. Scı. 33, 84. Best, T. L.; HART, E. B. (1976): Swimming ability of pocket gophers (Geomyidae). Texas J. Sci. 27, 361-366. CARTER, J. L.; MERRITT, J. F. (1981): Evaluation of swimming ability as a means of island invasion by small mammals in coastal Virginia. Ann. Carnegie Mus. 50, 31-46. Dacs, A. I.; WınDsor, D. E. (1972): Swimming in northern terrestrial mammals. Canadian J. Zool. 50, 117-130. ESHER, R. J.; WOLEE, J. L.; LAYNE, ]J. N. (1978): Swimming behavior of rice rats (Oryzomys palustris) and cotton rats (Sigmodon hispidus). J. Mammalogy 59, 551-558. Evans, R. L.; Katz, E.; Orson, N.; Dewssury, D. (1978): A comparative study of swimming behaviour in eight species of muroid rodents. Bull. Psych. Soc. 11, 168-170. FISHER, A. K. (1885): Ihe star-nosed mole amphibious. Amer. Nat. 19, 895. FIsHER, G. F. (1961): Behavior of salt-marsh Microtus during winter high tides. J. Mammalogy 42, 37-43. FooTs, L. E. (1941): A swimming hairy-tailed mole. J. Mammalogy 22, 452. FREDRICKSON, L. F. (1972): Swimming ability of Spermophilus richardsonii. J. Mammalogy 53, 190-191. GETZ, L. L. (1967): Responses of selected small mammals to water. Univ. Connecticut Occas. Papers Biol. Sci. Ser. 1, 71-81. HAFNER, J. C.; HAFNER, M. S. (1975): Water as a potential barrier to dispersal in Microdipodops. J. Mammalogy 56, 911-914. 20 G. C. Hickman HANAWALT, F. A. (1922): Habits of the common mole Scalopus aquaticus machrinus (Rafınesque). Ohio J. Sci. 22, 164-169. Harıs, C. E.; PETERSEn, M. K. (1979): Comparative swimming performances in selected cricetid and heteromyid rodents. Occas. Pap. Zool. 3, 1-16. Hıckman, G. C. (1977): Swimming behavior in representative species of the three genera of North American geomyids. Southwest. Nat. 21, 531-538. Hıckman, G. C. (1978): Reactions of Cryptomys hottentotus to water (Rodentia: Bathyergidae). Zoologica Africana 13, 319-328. Hıckman, G. C. (1979): The mammalıan tail: a review of functions. Mammal Rev. 9, 143-157. Hıckman, G. C. (1982): Climbing ability of the star-nosed mole, Condylura cristata (Talpidae). Säugetierkdl. Mitt. 40, 296-297. Hıckman, G. C. (1983a): Burrows, surface movement, and swimming of Tachyoryctes splendens (Rodentia: Rhizomyidae) during flood conditions in Kenya. J. Zool., London, 200, 71-82. Hıckman, G. C. (1983b): Swimming ability of a naked mole-rat (Heterocephalus glaber). Mammalıa 47, 267-269. Hıckman, G. C. (1984a): Influence of the semi-aquatic habit ın determining burrow structure of the star-nosed mole (Condylura cristata). Canadian J. Zool. 61, 1688-1692. Hıckman, G. C. (1984b): Swimming ability of talpıd moles with particular reference to the semi- aquatic Condylura cristata. Mammalıa 48, 505-513. Hıckman, G. C. (1984c): Behavior of North American geomyids during surface movement and construction of earth mounds. Special Publ. Mus. Texas Tech. Univ. 22, 165-186. Hıckman, G. C. (1985): Surface-mound formation by Ctenomys fulvus (Rodentia: Ctenomyidae) with comments on earth-pushing in other fossorial mammals. J. Zool., London, 205, 385-390. Hıckman, G. C. (1986): Swimming of Amblysomus hottentotus (Insectivora: Chrysochloridae) with notes on Chrysospalax and Eremitalpa. Cımbabesia, (A) 8, 55-61. HickmaAn, G. C.; MACHINE, C. (1987): Swimming behaviour in six species of South African rodents (Cricetidae, Muridae). Acta Theriologica 31, 447—464. Hıckman, G. C.; NEvo, E.; HETH, G. (1983): Geographic varıation in the swimming ability of Spalax ehrenbergi in Israel. J. Biogeogr. 10, 29-36. Howeıı, A. B. (1930): Aquatic mammals - their adaptations to life in water. Baltimore: Charles C. Thomas. Hury, L. M. (1941): Mammalıan invasıon via the highway. J. Mammalogy 22, 383-386. KENNERLYy, T. E., JR. (1963): Gene flow and swimming ability in the pocket gopher. Southwest. Nat. 8, 85-88. Kınc, J. A. (1961): Swimming and reaction to electric shock in two subspecies of doormice, Peromyscus maniculatus, during development. Anım. Behav. 9, 142-150. McCuutegy, H. (1967): The broad-handed mole, Scapanus latimanus, in a marine littoral environ- ment. J. Mammalogy 48, 480-482. MERRIAM, C. H. (1884): The star-nosed mole amphibious. Science 4, 429. MiLLER, M. A. (1957): Burrows of the Sacramental Valley pocket gopher in flood-irrigated alfalfa fields. Hilgardıa 26, 431-452. MooRE, A. W. (1939): Notes on the Townsend Mole. J. Mammalogy 20, 499-501. Nevo, E. (1979): Adaptive convergence and divergence of subterranean mammals. Ann. Rev. Ecol. Syst. 10, 269-308. Pıne, R. H. (1975): Star-nosed mole eaten by bullfrog. Mammalıa 39, 713-714. REED, C. A.; Rıney, T. (1943): Swimming, feeding, and locomotion of a captive mole. Amer. Midl. Natur. 39, 790-791. SAVAGE, 1. R. (1973): A stream as a barrier to homing in Peromyscus leucopus. J. Mammalogy 54, 982-984. SCHAPIRO, $. M.; SaLas, M.; VUKOVITCH, K. (1970): Hormonal effects on ontogeny of swimming ability in the rat: assessment of the central nervous system development. Science 168, 147-151. SCHMIDLY, D. J.; PACKARD, R. L. (1967): Swimming ability in pocket mice. Southwest. Nat. 12, 469-487. SMITH, M. F.; ParTon, J. L. (1980): Relationships of pocket gopher (Thomomys bottae) populations of the lower Colorado River. J. Mammalogy 61, 681-696. STARRET, A.; FISHER, G. F. (1970): Aquatic adaptations of the water-mouse, Rheomys underwoodi. Contrib. Sci. (Los Ang.) 182, 14. STOCK, A. D. (1972): Swimming ability in kangaroo rats. Southwest. Nat. 17, 98-99. VLECK, D. (1979): The energy cost of burrowing by the pocket gopher Thomomys bottae. Physiol. Zool. 52, 122-136. WALKER, E. P. (1975): Mammals of the World. Baltimore: John Hopkins Univ. Press. WILBER, C. G. (1958): Biological sımilarity and swimming time in rodents. Anat. Rec. 131, 608-609. WILBER, C. G. (1959): Some factors which are correlated with swimming capacity in guinea pigs. J. Appl. Physiol. 14, 199-203. WILBER, C. G. (1963): Te swimming capacity of the golden hamster. J. Mammalogy 44, 265-267. The swimming abılıty of Ctenomys fulvus and Spalacopus cyanus 21 WILBER, C. G.; Hunn, J. B. (1960): Swimming of albino mice. J. Appl. Physiol. 15, 704-705. Wırkıns, K. T. (1985): Variation in the Southeastern Pocket Gopher, Geomys pinetis, along the St. Johns River in Florida. Amer. Midl. Nat. 114, 125-134. Wırrıams, S. L. (1976): The effects of floods on Thomomys bottae in Texas. Southwest. Natur. 21, 169-175. WILBER, C. G.; WEIDENBACHER, G. H. (1961): Swimming capacıty of some wild mammals. J. Mammalogy 42, 428429. Author’s address: Dr. GRAHAM C. HıckMmAn, Department of Zoology and Entomology, University of Natal, P.O. Box 375, Pietermaritzburg, 3200 Natal, R.S.A. Z. Säugetierkunde 53 (1988) 22-30 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Wild boar social groupings and their seasonal changes in the Camargue, southern France By MARYSE DARDAILLON Centre de Recherches en Biologie du Comportement, Toulouse, France Receipt of Ms. 29. 1. 1987 Abstract Wild boar social organization and its monthly changes were studied from 872 sightings of groups, recorded in the Camargue (southern France) from December 1975 through February 1983. Groups, comprised of 1 to 23 anımals, were classified according to their composition. There were three recognized age categories: piglets, subadults, and adults. Of the 872 groups, 240 (27.5 %) were single adults, 134 (15.4 %) consisted of adults grouped together and 119 (13.6 %) were subadults observed either singly or grouped together. The most frequent groups (N = 379, 43.5 %) were matriarchal units; they were generally comprised of 1 to 5 sows with their piglets and/or yearlings. The study of monthly variations in the percentage frequency observed for the various group categories showed that the socıal structure of wild boar is a dynamic system that changes throughout the year in relation to the farrowing and breeding seasons. Introduction Despite the large literature on wild pıgs (MAUGET et al. 1984), there is little information on their grouping patterns. Group size and composition have been described in some studies dealing with the general ecology and behaviour of the species (SLupskrı 1956; SNETHLAGE 1957; GUND- LACH 1968; EISENBERG and LOCKHARDT 1972; FrRÄDRICH 1974; BEUERLE 1975; GRAVES and GraveEs 1977; BARRETT 1978; SHAFFER 1979; SANTIAPILLAI and CHAMBERS 1980; SINGER and ACKERMAN 1981; MEYNHARDT 1982). Formation of family units appears characteristic of wild pigs. Their social organization has been described as a matriarchal society with the basic famıly group comprised of a sow and its young of the year. Multi- family groups, composed of several sows with their young have been termed “sounders” by EisEnBERG and LOcKHARDT (1972). Extended family groups also include subadults which are presumably offspring of the previous year (SHAFFER 1979). In contrast, adult males are solitary; they join females only during the breeding season. Little attention has so far been paid to seasonal changes in grouping patterns and their determining factors. SLuDskIr (1956) or SANTIAPILLAI and CHAMBERS (1980) have shown that various environmental factors such as temperature, snow cover or availability of food can cause changes in group size of wild pigs. In other respects, various results indicate that changes ın social organization occur during rutting and farrowing time. The present paper reports a study of group size and composition of wild boars living ın the Camargue, and analyzes changes related to the farrowing and breeding seasons. Study area Sıghtings of wild boar groups have been recorded at the Tour du Valat Reserve, which is privately owned land with minimal human access. It covers an area of 1500 ha and is located in the southeastern part of the Rhöne delta. Lying to the east of the Camargue National Reserve, the area is surrounded on its other sides either by cultivated land or by more or less natural ground. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5301-0022 $ 02.50/0 Wild boar social groupings and their seasonal changes 23 300 100 IL] 1-5 ) = I—) s s 50 L-} E = = ı 280 A gez SEEOETOREN 12213714 152 16 172218019 7207721722 723 Group size Fıg. 1. Frequency diagram indicating the group size for the wild boar in the Camargue, southern France Material and methods Observations and sightings of wild boar groups were mainly made by game wardens and researchers of the “Tour du Valat Biological Station” during their regular field work. From December 1975 through February 1983, these observers noted the composition and location of the wild boar groups observed on pre-established tables and maps, which were fixed on a board in the hall of the research building. The anımals observed were classed in three age categories according both to their estimated weight and coat color. Piglets, i.e. young with a striped coat, were generally less than 4 months old and weighed less than 15-20 kg. Subadults had a reddish or black coat, and a weight varyıng from 15-20 to 30-40 kg. Adults referred to anımals more than 2 years old; they looked big, heavy and dark or silver grey colored. With a few exceptions, sex was not specified because it could not be reliably determined at a distance. Results A total of 3530 animals in 872 distinct sightings were observed during the 7 years of the study. Of the 3530 wild boars observed (fig. 1), 240 (8.2 %) were single adults, 59 (1.7 %) were single subadults or piglets, while 3241 (91.8 %) anımals were encountered in 583 groups composed of 2 through 23 anımals. The most frequent group sizes were 2 (18.0 %) 24 Maryse Dardaillon and 3 (16.3 %), then 4 and 5 (13.5 % each). Groups of more than 10 anımals were seen only occasıonally (12.3 % of the 583 groups). Group categories The 872 sightings ranged in the 11 following categories: 1. Single adults (n = 240, 27.6 %); 2. Adult-groups (n= 134, 15.4 %); 3. Single subadults (n= 41, 4.7 %); 4. Subadult- groups (n= 78, 8.9 %); 5. Groups of both adults and subadults (n= 281, 32.2 %); 6. Basic famıly group = 1 adult with piglets (n = 27, 3.1 %); 7. Sounders = several adults (2 to 5) with piglets (n = 36, 4.1 %); 8. Extended family groups = adults with both piglets and subadults (n = 16, 1.8 %); 9. Single piglets (n = 8, 1.0 %); 10. Piglet-groups (n = 9, 1.0 %); 11. Subadult- and piglet-groups (n = 2, 0.2 %). Categories numbered 5 to 11 must all be considered as matriarchal in nature. Groups of adults with piglets include the categories “family groups” (n° 6) and “sounders” (n° 7), previously mentioned, whereas groups comprised of adults and subadults (n° 5) are supposedly a later stage of this type of association. Category n? 8 has previously been termed “extended family group”. The 19 groups that make up the categories n° 9, 10 and 11 have to be included in matriarchal associations because they all had at least one piglet; we suppose that one or more sows were either absent or out of sight for the observer. Monthly variation in average group size The average group sıze was about 4 anımals. It ranged from 3.3 to 3.7 from January to May. It exceeded the annual average from June through December, ranging from 4.2 in July to 5.2 in October (fig. 2). 6 5— JFMAMAJJASOND Fıg. 2. Monthly variation in mean group size, in wild boars living in the Camargue, southern France A more precise analysıs has revealed that the proportion of groups formed of at least 4 anımals increased from July through December, whereas that of groups including less than 3 anımals decreased. The shift is observed only in August if the comparison takes into account groups of more than 5 anımals and those of less than 4 animals. This means that wild boars regroup gradually during the summer and autumn seasons, and consequently live ın larger groups during the second part of the year than during the first part. 85 rol 6€1 ROT, vr LE Se IC) saLLO331e9 [eıoagns sıofdıd Yam sıpnpegng [@) sdno13-10[314 N s29[31d a[durg sdno.3 Ajrure} popusıx N sI9punog sdno.3 Aprurey sıseg sıynpegns PLSSIIUDV N [ee] an 0 0 I 0) I rI [4 0) I 8 1% — un — x o oo sdno.13-ıpnpegns mn sıpnpegns af3urg © —_ sdno13-ı[npy S 2 sown--nivron a sıynpe af3uıg Kıo3a2ea dno1ny, („JueIg uay3nos) an3ıewen 94} ur PaA1asgo saLıo3ayeI dno13 1eoq pjim snorseA 3yJ Jo Aduanbaıy aanejas ayyJ ur saduey> Ajyyuop '27qPL 26 Maryse Dardaillon Monthly changes in group composition Changes concerning gronps with piglets 98 groups — categories n° 6 to 11 - included at least one piglet (table). Their percentage frequency greatly increased ın Aprıl (fig. 3) and reached its maximum in May (41.4 %, table) when most females had given birth; at that time piglets follow their dam and were thus more easıly sighted. Piglets were seen alone ın 11.1 %, 16.3 % and 58.3 % of the observations in March- April, May through June and August-September, respectively. This result indicates that as they increase % in age piglets spend relatively less time in the close 40 vicinity of their dam. ' Groups including piglets became rarer in August, then again in September when piglets changed their coat color and were therefore ıdentified as subadults by the observers (fig. 3). From October through March, the monthly fre- quency was low and more or less constant. A basic level, less than 5-6 % (table), was registered during this period, which means some births occurred every month. The maximal percentage frequency of basic family J groups was observed in April and May (fig. 4"), whereas that of sounders and extended family groups 10 was reached in May for the first category (fig. 4°) and in June for the second (fig. 4"). | m= 11.2 Changes concerning subadults Emalmis'stalsioin!o The percentage frequency of groups comprised of Fig. 3. Monthly variation in the rela- both adults and subadults greatly diminished in April tive frequency of wild boar groups (-29.2 %, table). At the same time, sightings of including at least one piglet (N = 98) single subadults and of subadult-groups increased. respectively from 2.2 to 12.5 % and from 10.1 to 13.5 % (table). These varıations are related to changes in social organization which happen during the prefarrowing period. At that time, sows become separated from their congeners and thus from subadults. Single subadults were most frequently observed in April (fig. 4°). They certainly regrouped during the following weeks as indicated by comparison of figures 4 and 4". We can thus observe that the percentage frequency of single subadults diminished in May and June, whereas that of subadult-groups became higher than in April. Groups comprised of females with both piglets and subadults were formed again during May through July (fig. 4"?). We therefore observed that the frequency of subadult-groups clearly decreased from May through August (fig. 4") whereas groups of adults with subadults (fig. 4°) increased by 44.4 %. The increase was particularly high (+32.4 %, table) between July and August because piglets, which had lost their stripes, were subsequently identified as subadults by the observers. Changes concerning adults Sightings of adults observed either alone or grouped together (fig. 4°") decreased progressively from April through August (respectively -11.8 % and -5.8%, table), on % 40 20 10 Wild boar social groupings and their seasonal changes 27 Single adults A % Adults with subadults © 50 Adult - groups B m 154 40 m= 32.2 30 JFMAMJJASOND JFMAMJJASOND 20 ig subadults D Subadult - groups E | m=89 10 10) JFMAMJ JASOND JFMAMJJASOND JFMAMJJASOND Basic family groups F Sounders G Extended family groups H m 41 JFMAMJJASOND JFMAMJJASOND JFMAMJJASOND Fig. 4. Monthly varıation in the relative frequency of the various group categories, in wild boars living in the Camargue, southern France. In the fig. 4° black parts indicate relative frequency for adult- groups comprised of at least 5 anımals account first of the birth of piglets then of the reformation of extended family groups. From August through March, adults were encountered mainly in the company of subadults (table). However, adult-groups were rather frequent from October through February (fig. 4°). Adults appeared to regroup in September (+3.6 %) then in October (+7.0 %). Ihe percentage frequency of groups including at least 5 adults was clearly above the average annual percentage from October through December (fig. 4°). This regrouping 28 Maryse Dardaillon of adults was related to the onset of the breeding season which was at its peak in early December. Afterwards, some adults, especially male wild boars, became solitary again as attested by the increase (+10.4 %, table) in the percentage frequency of single adults in January (fig. 4°). At that time, adult-groups remained rather frequent (fig. 4°), but they were smaller than observed during the prerut and rut seasons. Discussion Wild boars living ın the Camargue exhibit the same basic social organization as has been previously described in other areas. Adults live either singly or in groups almost always including juveniles (subadults and/or piglets). According to the literature and our own observations of marked individuals (unpublished data), we think that single wild boars are mainly males whereas those living in groups are sows. When not in the company of adults, subadults are encountered either alone or grouped together. Our results show that the socıal structure of wild boar is a dynamic system that changes throughout the year in relation to the farrowing and breeding seasons. At farrowing time, i.e. mainly in April, matriarchal groups are disrupted when gestating sows become isolated some days before they give birth. MEYNHARDT (1982) observed a 4-year female at farrowing time and reported that the female was followed by its young of the previous year when separating from congeners; afterwards, young remained with their mother which kept them at a distance of at least 20 m during nest building, farrowing and the post- partum period. However, the separation from congeners seems to be the rule in most other areas (SNETHLAGE 1957; GUNDLACH 1968; FrÄDRICH 1974). In addition, a number of radiotracking studies have shown that sows reduce their movements and tend to center their activities around nests before farrowing (Kurz and MARCHINTON 1972; BARRETT 1978; Mauer 1980; DouAaupD 1983; JanEAU and Sprtz 1984). Consequently, subadults remain alone or grouped in small units. Results obtained in the Camargue show that single subadults would soon attempt to find companions. After a postpartum seclusion period (Kurz and MARCHINTON 1972; MAUGET 1980; SINGER et al. 1981), piglets leave the farrowing nest and follow their mother. 'The family group, comprised of the sow with its piglets, may thereafter grow larger by association with other anımals as has been observed by SNETHLAGE (1957), GUNDLACH (1968), MAucET (1980) or in the present study. Our results show that sounders — which result from the joining up of at least two family groups - are chronologically the first type of association. The good synchronization of births observed in the Camargue would facilitate their formation since piglets are of similar age and development. In other areas, sounders have been observed to form during the first two weeks after birth (SNETHLAGE 1957; MEYNHARDT 1982). Sounders are the equivalent of nursery groups reported in other species of Ungulates such as carıbou (Prurtt 1960), bighorn sheep (GEIST 1971) or reindeer (HELLE 1981). However, these groups seem to be smaller in wild pigs, although large sounder assocıiations have sometimes has been observed by several authors (SNETHLAGE 1957; MEYNHARDT 1982; other references in SINGER and ACKERMAN 1981). Extended family groups are formed from May through June. As observed in bighorn sheep (Geist 1971), it seems that reassociations with yearlings occur only 34 wecks after young are born. For SNETHLAGE (1957), the spring separation from the mother is definitive. If some yearlings rejoin sounders in August, i.e. 4-5 months after births, most of them remain together and form large subadult-groups (“companies”). For hıs part, MEYNHARDT (1982) observed that the anımals of the same matriarchal unit regroup only a few days after farrowing. In this case, the separation of group members is both more limited (see above) and more temporary than observed in the Camargue. Wild boar social gronpings and their seasonal changes 2,9 Strong bonds exist between family members and especially between closely related females (mother-daughter, sisters...) which would preferentially reassociate after the farrowing seclusion. FRÄDRICH (1974) indicates that young females live with their mother until they reach sexual maturity. However, even if it lasts only a short time, the disassociation is certainly favourable to the dispersal of yearling temales. They may follow some older female and thus switch to another matriarchal group if they happen to meet one. With regard to the disassociation of yearlings from their mother during the rutting season, results obtained by various authors are also quite different. Most authors state that yearlıng males are chased away by adult boars (NasımovicH 1966 ın FRÄDRICH 1974; BEUERLE 1975; SANTIAPILLAT and CHAMBERS 1980). With respect to 2 years old males, which reach their sexual maturity at that time, SNETHLAGE (1957) indicates that they leave the group voluntarily. For GuUnDLAcCH (1968), female yearlings are also chased away by boars. This author states that both male and female yearlings rejoin their mother after rut. For hıs part, BEUERLE (1975) thinks that the disassociation of yearling males is definitive; they would live in subadult-groups, then become solitary. In the Camargue, wild boars are more gregarious from June through December than during the first part of the year. The groups grow larger firstly on account of the formation of sounders and extended family groups, then, during the prerut and rut periods, because of boars joining matriarchal groups to breed. The prerut period is characterızed by an increase of adult-groups including at least 5 anımals, from October through December. This prerut period would also exist in other populations of wild pigs. For example, PFEFFER (1961) indicates that SLupsk1t (1956) observed fighting between adult boars from August while in Central Asıa breeding activities occur later on, in November and December. Acknowledgements The research was supported by the Foundation Tour du Valat. I am grateful to all people who have placed their observations of wild boars at my disposal. I wish to thank Prof. R. Campan for his advice and guidance during the course of the study, J. SMITH and R. Cooke for correcting the English translation, and Y. MEYER for providing the German version of the summary. Zusammenfassung Soziale Gruppenbildungen und ihre jahreszeitlichen Änderungen bei Wildschweinen in der Camargue, Südfrankreich 872 Gruppen von Wildschweinen, die zwischen 1 und 23 Individuen enthielten, konnten in der Camargue beobachtet werden. Sie wurden aufgrund ihrer Altersverteilung klassifiziert. 240 (27,5 %) der beobachteten Gruppen bestanden aus einzelnen Adulten, 134 (15,4 %) enthielten mehrere Individuen, jedoch nur Erwachsene, und 119 (13,6 %) waren entweder einzelne Jungtiere oder Jungtiergruppen. Die häufigste Gruppenbildung (379, 43,5 %) stellten Mutterverbände dar. Sie bestanden aus 1 bis 5 Bachen, die mehrere Frischlinge und/oder Jungtiere führten. Die Analyse der monatlichen Änderungen i in der Häufigkeit der verschiedenen Gruppen zeigt, daß die soziale Struktur bei Wildschweinen ein dynamisches System ist, dessen Änderungen im Laufe des Jahres mit der Paarungszeit und der Geburtszeit korreliert sind. References BARRETT, R. H. (1978): The feral hog on the Dye Creek Ranch, Calıfornia. Hilgardıa 46, 283-355. BEUERLE, W. (1975): Freilanduntersuchungen zum Kampf- und Sexualverhalten des europäischen Wildschweines (Sus scrofa L.). Z. Tierpsychol. 39, 211-258. DouvaAup, J.-F. (1983): Utilisation de l’espace et du temps et ses facteurs de modulation chez le sanglier, Sus scrofa L., en milieu forestier ouvert (Massıf des Dhuits, Haute-Marne). These de Doctorat de 3eme cycle, Univ. Strasbourg. EISENBERG, J. F.; LOCKHART, M. (1972): An ecological reconnaissance of Willpattu National Park. Washington, D.C.: Smithsonian Inst. Press. 30 Maryse Dardaillon FRÄDRICH, H. (1974): A comparison of behaviour in the Suidae. In: The behaviour of Ungulates and its relation to management. Ed. by V. Geist and F. WALTHER. IUCN Publ. New Series 24, 133-143. Geist, V. (1971): Mountain sheep. A study in behavior and evolution. Ed. by G. B. SCHALLER. Chicago and London: Univ. of Chicago Press. Gravss, H. B.; GrAvEs, K. L. (1977): Some observations on biobehavioral adaptations of swine. In: Research and Management of wild hog populations. Proc. of a Symp. Ed. by G. W. Woon. Georgetown, $.C.: Belle W. Baruch Forest Science Institute of Clemson University; 103-110. GUNDLACH, H. (1968): Brutfürsorge, Brutpflege, Verhaltensontogenese und Tagesperiodik beim Europäischen Wildschwein (Sus scrofa L.). Z. Tierpsychol. 25, 955-995. Here, T. (1981): Observations of home ranges and grouping patterns of the free-ranging semi- domestic reindeer (Rangifer tarandus tarandus L.) ın Kuusamo, northeastern Finland. Research Institute of Northern Finland, Univ. Oulu A 2, 34-54. JanEAU, G.; Spitz, F. (1984): L’espace chez le sanglier (Sus scrofa scrofa L.). Occupation et mode d’utilisation. Gibier Faune Sauvage 1, 73-89. Kurz, J. C.; MARCHINTON, R. L. (1972): Radiotelemetry studies of feral hogs in South Carolina. ]. Wildl. Manage. 36, 1240-1248. MaAucGET, R. (1980): Regulations Ecologiques, comportementales et physiologiques (fonction de reproduction) de l’adaptation du sanglıer, Sus scrofa L., au milieu. These de Doctorat d’Etat, Univ. Francois Rabelais, Tours. MAUGET, R.; CAaMPANn, R.; SPITZ, F.; DARDAILLON, M.; JANEAU, G.; Pepın, D. (1984): Synthese des connaissances actuelles sur la biologie du sanglier, perspectives de recherche. In: Symposium international sur le sanglier, Toulouse (France). Ed. by INRA Publ. Coll. INRA 22, 15-50. MEYNHARDT, H. (1982): Schwarzwild-Report. Mein Leben unter Wildschweinen. Leipzig: Neumann Verlag. ran (1961): L’ecologie du sanglier en Asie centrale d’apres les recherches d’A. A. SLOUDskY. Rev. Ecol. (Terre Vie) 108, 368-372. PrurTT, W. ©. (1960): Behaviour of the barren-ground carıbou. Biol. Papers Univ. Alaska 3, 144. SANTIAPILLAI, C.; CHAMBERS, M. R. (1980): Aspects of the population dynamics of the wild pig (Ss scrofa Linnaeus, 1758) ın the Ruhuna National Park, Srı Lanka. Spixiana 3, 239-250. SHAFFER, M.L. (1979): Behavior of the European wild boar in the Great Smoky Mountains National Park. Proc. 1st Conf. Scı. Res. in the National Parks. National Park Serv. Trans. and Proc. Ser. 5, 357-363. SINGER, F. J.; ACKERMAN, B. B. (1981): Food availability, reproduction, and condition of the European wild boar ın Great Smoky Mountains National Park. U.S. Depart. Int., Natl. Park Serv., NPS-SER Research/Resources Management Report 43, 1-52. SINGER, F. ]J.; OTTo, D. K.; Tıpron, A. R.; HAsLe, C. P. (1981): Home ranges, movements, and habitat use of European wıld boar in Tennessee. J. Wildl. Manage. 45, 343-353. Srupsku, A. A. (1956): Kaban, morfologiya, ekologiya, khozyaistvennoe i epizootologicheskoe zhachenie. Dromyscl. Alma Ata. Akademii Nauk Kazachskoı SSR. SNETHLAGE, K. (1957): Le sanglier. Histoire naturelle et chasse. (Traduction H. MANHES D’ÄNGENY). Metz, France: Editions Le Lorrain (1977). Author’s address: Dr. MArysE DARDAILLON, Centre de Recherches en Biologie du Comportement, Universite Paul Sabatier, U.A. n° 664 CNRS, 118 Route de Narbonne, F-31062 Toulouse Cedex, France Z. Säugetierkunde 53 (1988) 3147 © 1988 Verlag Paul Parey, Hanburg und Berlin ISSN 0044-3468 Variabilite comportementale a l’interieur du genre Cephalophus (Ruminantia, Bovidae), par ’exemple de C. rufilatus Gray, 1846 Par G. DusosTt et F. FEER Laboratoire d’Ecologie Generale du Museum, Brunoy, France Reception du Ms. 27. 10. 1986 Abstract Behavioural differences in the genus Cephalophus (Ruminantia, Bovidae), as illustrated by €. rufılatus Gray, 1846 Studied the behaviour of Cephalophus rufilatus in captivity. Although most behavioural characters are shared by other duikers, this species shows some unique features as yet not encountered in the genus Cephalophus: lack of social play and reciprocal allogrooming, Laufschlag very rudimentary, mount of partners by the receptive female, droppings in pile, etc. ... C. rufilatus is intermediate between diurnal and nocturnal species, according to specific behaviour as well as social life. The genus Cephalophus ıs very homogeneous in basic behaviours such as locomotion, excretion, marking and reproduction, but variable, considering behavioural patterns directly depending on life conditions (habitat, diet and activity rhythm). Accordingly, due to actual presence of “archaic”’ and “advanced” characters, the duiker group cannot any longer be considered as the most primitive among Bovids. Introduction L’etude de la biologie des Cephalophes connait actuellement un certain developpement, et ceci correspond sans doute au fait que cette sous-famille de Bovides est morphologique- ment l’une des plus homogenes, qu’elle est abondamment representee dans toutes les formations forestieres d’Afrique, et que son Etude E£tait restee jusqu’a present A peu pres totalement negligee. Faute d’arguments paleontologiques, on ne connait rien de l’origine de ce groupe original, de ses rapports avec les autres Bovides et ıl est donc ditficile de lui assigner un niveau phylogenetique et Evolutif precıs. Dans la plupart des ouvrages, les Cephalophes sont habituellement presentes selon un modele unique de petit ruminant forestier, cr&pusculaire ou nocturne, et solitaire. Cela reflete l’extreme indigence des renseignements dont on dispose sur eux, mais cela re- presente egalement une carıcature de la realıte, puisqu’il existe, au contraire, de grandes differences interspecifiques: — dans la taille et le poids corporel (3 ä 80 kg); — dans la morphologie (especes pourvues de cornes dans les 2 sexes ou, au contraire, cornes presentes uniquement chez les mäles; corps lourd et trapu, monte sur des pattes courtes, ou leger avec de longs membres); — dans l’Ecologie (hötes de pleine for£t, de regions arbustives ou m&me de steppes; especes de terrain sec ou especes localısees au bord des cours d’eau); — dans l’alimentation (frugivores ou folivore-herbivores); — dans le rythme d’activite (diurnes ou nocturnes exclusifs, ou dıurne-nocturnes); — dans le comportement et la vie sociale (solitaires ou couples stables; contacts interindividuels rares ou tres frequents et varies). Une espece, Cephalophus rufılatus Gray 1846, semble occuper une place particuliere, puisqu’elle presente un certain nombre de traits permettant de relier les nombreuses especes de pleine foret du genre Cephalophus au genre monospecifique de savane et de steppe Sylvicapra. Comme la plupart des autres especes de Cephalophes, on peut estimer que sa biologie est inconnue, puisque seul DITTRICH (1972) en a fait une breve &tude en U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5301-0031 $ 02.50/0 32 G. Dubost et F. Feer captivite. Par comparaison avec les quelques especes sur lesquelles nous avons quelques renseignements precis, ıl semble donc necessaire de dresser un premier biılan de son comportement, de maniere A pouvoir mieux comprendre le groupe des Cephalophes par comparaison avec les autres Bovides. C. rufilatus se rencontre au nord de l’Equateur, de la Gambie jusqu’ä l’ouest de l’Ouganda. C’est un habitant des forets claires, galeries forestieres ou bordures de foret, parfois m&me de la savane, mais jamais de la foret ombrophile dense. C’est une espece d’assez petite taille, de 10 ä 14 kg (BrLancou 1958, 1962; HALTENORTH 1963; KINGDON 1979, 1982; Ras 1984). Le corps est legerement bas du devant, de couleur rouge-orangee avec une teinte gris-noir sur le museau et le front, le milieu de la croupe, la queue et les pattes. Iln’y a pas de plages blanches bien nettes, sı ce n’est un peu sur la levre sup£rieure et le bord des oreilles. La touffe frontale est developpee dans les deux sexes. Les glandes preorbitaires, pro@minentes, s’ouvrent selon une ligne rectiligne de pores. L’espece est pourvue de glandes inguinales. Le mäle seul possede des cornes, la femelle ne presentant que des petits boutons osseux plus ou moins invisibles ä l’exterieur. On ne connait nı son rythme d’activite nı son regıme alimentaire: elle pourrait Etre diurne-nocturne et se nourrir de fruits, feuilles et herbes (BAUDEnoN 1958; HALTENORTH 1963; KınGDon 1982). Dans la nature, on rencontre les anımaux le plus souvent seuls, rarement ä 2 ou 3. Materiel et methodes Les observations ont £t£ faites au Zoo de Pietat (France) en septembre 1982 et juin 1983. Les animaux avaient ä leur disposition un terrain d’environ 7.000 m?, comprenant une prairie centrale de 2.800 m?, parseme&e de quelques arbres et bordee par une zone partiellement boisee comprenant differents abrıs. L’espece y a Et& introduite en 1974. Depuis cette date, des naissances ont eu lieu regulierement chaque ann&e. Au debut de notre Etude, la population comptait 6 individus: un mäle adulte de 6 ans (M1), 2 femelles adultes de plus de 7 ans (Fl et F2), une femelle subadulte de 2 ans (F3), un jeune mäle de 18 mois (M2) et une jeune femelle de 7 semaines (F4). La femelle F1 etait la mere de MI, M2, F2 et F4. La femelle F2 etait la mere de F3; elle donna naissance A un nouveau jeune en juin 1983. Comme le suppose DITTRICH (1972), les naissances ont lieu tous les mois de l’annee (janvier: 1; fevrier: 1; mars: 1; mai: 1; juin: 1; aoüt: 1; septembre: 2; decembre: 1). Une duree de gestation de 248 jours a ete calcul&e pour la femelle F2, ce qui correspond aux chiffres fournis par DITTRICH (223-241 jours). Comme chez C. zebra (SCHWEERS 1984), la femelle peut se retrouver en oestrus tres peu de temps apres avoir mis bas (6 jours chez F2), ce qui correspond vraisemblablement a un oestrus post-partum bien plus pr&coce que les 4,5 semaines citees par DITTRICH. Les observations ont Et€ conduites en suivant ä tour de röle chaque individu durant une journee complete, exception faite de la longue periode de repos de la mi-journee. Durant chaque sejour, chaque anımal a &t€ ainsi observ& pendant 2 journees espac&es l’une de l’autre par 4 jours. Tous les comportements individuels et sociaux ont Ete decrits, photographies, comptes et chronometres. Resultats Description des differents comportements Activite Dans les conditions de captivite, C. rufilatus apparait comme une espece essentiellement diurne. Lors des visites que nous avons faites au cours de la premiere moitie de la nuit, tous les anımaux &taient inactifs, m&me par claır de lune; s’ıls etaient deranges par la lumiere de la lampe, ils se levaient, mais se recouchaient peu de temps apres. Les maximums d’activit€ se produisent le matin et en fin d’apres-midi, avec une pause plus ou moins absolue en milieu de journee. Chaque animal est generalement actıf A 45-70 % de son temps, et ıl n’apparait pas de differences individuelles notoires, exception faite des jeunes qui sont beaucoup moins actifs que les subadultes ou adultes. Les repos qui sont pris en cours de journee se passent souvent sous abri, mais ıls Comportement de C. rufilatus et variabilite du genre Cephalophus 33 peuvent €galement avoir lieu en plein air et ä decouvert: milieu de prairie ou pied d’un arbre ou d’un rocher. On compte 5 ä 10 places de repos diurne differentes par individu; 1 & 3 d’entre elles sont plus utilisees que d’autres et ce sont alors des places abritees. Les jeunes sont particulierement fideles ä une place determinee. Contrairement ä la journee, les places de repos nocturne sont peu nombreuses; ce sont toujours des places abritees et retirees. Il n’y a aucune place de repos propre ä un individu, pas plus qu’il n’existe un quelconque partage du terrain ou une defense d’une zone par un individu quelconque, m&me lorsqu’il y a un nouveau-ne. De jour comme de nuit, le regroupement de 2 ou plusieurs individus dans la m&me place de repos est donc courante. Deplacements L’animal marche la tete et le cou portes ä hauteur de garrot; la queue est alors continuelle- ment agitee d’un cöte A l’autre. Parfois, il hume en relevant le museau & l’oblique. L’inquietude se manifeste par un soufflement qui est r&pete. L’anımal s’arrete, la tete droite, une patte anterieure ou posterieure bloquee en l’air. Une reconnaissance de la source d’inquietude peut se faire & 10 m de distance, selon un deplacement en demi-cercle, accompagne& de soufflements, mais sans tapement du sol. La fuite se faıt par une serie de bonds allonges, les 2 pattes anterieures puis les 2 posterieures & l’unisson. C. rufilatus ne possede aucun cer&monial de coucher; ä une seule occasion, un individu a brievement gratt@ la place. Par contre, tous les individus peuvent gratter le sol de plusieurs coups de leurs pattes anterieures, alternativement ou non, pour chercher & manger. Pour prendre des bourgeons et des feuilles au bout des branches, ıls se dressent sur leurs pattes arrıeres, mais ils ne sont pas capables de maintenir longtemps cette position. Le jeu est presque exclusivement le propre des jeunes et subadultes. Il est toujours solitaire et principalement a base de course et de saut: l’anımal court autour d’un adulte ou d’un obstacle, en effectuant des demi-tours, des sauts sur place ou des bonds allonges. Miction, defecation, Flehmen Iln’y a pas d’attitude corporelle notable lors de la defecation ou les anımaux s’accroupis- sent tres peu, sans avoir ni gratt€ nı marque la place, mais souvent apres avoir senti le sol. Par contre, pour la miction, la femelle baisse son arriere-train presque jusqu’au sol et le mäle recule fortement les pattes arrieres. L’urine et les feces ne sont jamais mis d’une maniere dispersee, mais au contraire presque tou- jours deposes sur certains emplacements situes sur une surface degagee comme la prairie. Chaque nou- velle Emission de feces se fait donc sur une des zones deja utilisees A cet effet: 98 % des laissees (342/350) se rencontrent ainsı sur moins de 11 % de la surface de la prairie. Chaque zone represente de 10 A 40 m?, et comprend de 10 A 70 laissees differentes. Ei ern en dukmalenibouche peu Le Flehmen est uniquement le fait du mäle adulte entrouverte (dessin d’apres photo) a l’urine d’une femelle. Au cours de ce comporte- ment, l’attitude corporelle est tout & fait semblable a celle des autres cephalophes: tete droite ou tourn&e, bouche entrouverte (Fig. 1). Le Flehmen est presque toujours unique et dure peu longtemps: 9,2 secondes # 2,0 (N = 16). 34 G. Dubost et F. Feer Marquage preorbitaire Comme tous les Cephalophes, le marquage preorbitaire est un comportement frequent chez tous les individus qui dispersent ainsi partout ou ıls passent leurs secretions sur tous les supports verticaux, A hauteur de tete. L’individu flaire tout d’abord ä& petits coups la place a marquer, tete droite, puis il y applique sa zone glandulaire en tenant la tete de cötE et ıl la frotte en ouvrant la bouche et en tirant souvent la langue (Fig. 2). Au lieu de flairer la place, il peut eEgale- ment la lecher ou la gratter avec ses incisives inferieures. Plusieurs marquages peuvent etre faits successivement par un m&me ani- mal a la m&me place, mais ıl y a toujours un contröle du support entre 2 marquages suc- cessifs. Les adultes peuvent parfois flairer ou lecher une place sans la marquer, mais ce comportement est plutöt le propre des jeunes qui ne semblent pas marquer avant la 2e annee. 81 & 99 % des marquages sont uniques (N = 2887), mais les femelles peuvent repeter leurs marquages plusieurs foıs (2 A 6) sur le m&me site, les mäles le faisant rarement plus de 2 fois. Ces marquages r&petes peuvent se faire en utilisant alternativement les 2 cötes de la tete, et A des places differentes. Lorsque 2 individus se trouvent en activiteE au m&me endroit, ıl arrıve que le comportement de marquage de l’un incite le 2e ä venir marquer aussitöt apres sur le m&me site ou A cöt£&; cela est particulierement Evident lorsque les 2 aniımaux sont engages dans un comportement agonistique. Le marquage pr£eorbitaire a lieu durant toutes les periodes actıves de l’individu; il est d’autant plus frequent que l’anımal est excite. On en compte en moyenne de 21 & 60 par heure d’activite, mais les femelles semblent un peu plus actives ä ce point de vue que les mäles (respectivement 25 ä 60 marquages par h. contre 21 A 34), mais cela peut correspon- dre dans cet Elevage A un niveau de competition plus grand entre elles qu’entre les mäles. Dans les 2 sexes, la frequence des marquages croit directement avec l’äge. Sur ce petit terrain d’environ 0,7 ha, chaque individu utilise de 107 & 175 sites de marquage differents. Lä encore, les femelles possedent plus de sites que les mäles et, chez elles, le nombre varıe conformement & l’äge; ıl en est de möme du nombre moyen de marquages sur chaque sıte: 2,09 a 3,14. Le degre d’utilisation des differentes portions du terrain pour le marquage preorbitaire est sensiblement le m&me pour tous les individus: il n’y a pas de zone tres marqu&e par un anımal et peu marquee par d’autres. Puisqu’il n’y a pas de partage du terrain entre les divers individus, la grande majorite des sites de marquage sont utilises en commun. On ne compte que 15 & 30 % des sites qui soient propres & un seul animal et, dans ce cas, leur frequentation par ce dernier est nettement inferieure & celle des sites communs: 7 4 17 % seulement du total des marquages effectues, soit en moyenne 0,77 & 1,74 par site. Un site utilise par plusieurs anımaux est donc marqu& par chaque individu 1,5 a 3,8 foıs plus qu’un site propre. Le nombre de marquages par un anımal sur un site donne est de m&me plus eleve lorsque ce site est utilise par plusieurs individus que s’il n’est frequente que par 2 d’entre eux (2 & 3,5 foıs plus). La frequence d’utilisation d’un site de marquage par un individu depend donc directement de celle des autres anımaux. Globalement, le mäle M1 apparait tres stimul& par les sites de marquage des autres anımaux: ıl y marque lui-m&me 3,1 3,8 fois plus que sur ses propres sites; la jeune femelle ‚Fig. 2. Marquage preorbitaire d’une femelle contre un tronc (dessin d’apres photo) Comportement de C. rufilatus et variabılite du genre Cephalophus 35 F3, par contre, est la moins stimulee par les sites de marquage des autres: elle n’y marque que 1,5 3 1,7 fois plus que sur ses propres sıtes. Cependant, les sites de F3 sont ceux qui induisent le plus grand nombre de marquages chez les autres anımaux, et ceux du mäle Mi le moins; les sites des femelles adultes F1 et F2 sont intermediaires. Lorsque 2 individus ont en commun, exclusivement entre eux deux, un certain nombre de sites de marquage, le nombre de ces derniers ainsi que l’intensit€ des marquages effectu&s sur chacun d’eux dependent directement de l’identitE des protagonistes. La femelle F1 montre ainsi une forte attirance pour les sites de marquage de la femelle F2; inversement, F2, F3 et M1 montrent tous une preference pour ceux de Fl. Marquage avec les cornes Chez C. rufilatus, le mäle est le seul sexe A &tre pourvu de cornes. Cependant, contraire- ment aux autres especes de Cephalophes, il marque tres rarement avec elles: un seul cas observ& sur 128 h d’observation, au cours duquel le mäle avait räcle ses cornes 2 foıs de suite contre le bord d’un abri, sans manıfester aucun autre comportement annexe, comme un grattage du sol. Il en est probablement de m&me des femelles, bien que ces dernieres puissent frotter parfoıs leur front et les petits boutons osseux contre differents supports horızontaux ou verticaux, en tournant la tete d’un cöte a l’autre; ce comportement n’a ete observe que chez la femelle la plus ägee F1 (6 observations), plus rarement chez F2 (2 observations) et jamais chez les autres. Allogrooming et marquage du partenaire En dehors des interactions sexuelles ou agonistiques, les contacts corporels entre 2 in- dividus sont de 3 ordres: contröle olfactif, lechage ou marquage du partenaire. Deux animaux peuvent se sentir la region nasale, soıit que l’un d’eux se soit approche de son partenaire, soit que les deux aıient effectue simultanement ce comportement (32 % des 245 contacts observes). Mais le contröle olfactif du partenaire de loın le plus frequent se fait dans la region ano-genitale (55 % des contacts), beaucoup plus rarement sur une autre partie du corps comme le garrot, les flancs ou les pattes (13 % des contacts). Les contröles olfactifs sont toujours brefs; ils sont egalement peu frequents, puisqu’ils ne concluent que 20 27 % des rapprochements entre individus. Le lechage est encore plus rare: on en compte en general 2 foıs moins que de contröles olfactits, en dehors des sequences de poursuite sexuelle ou de comportement maternel. Le lechage du partenaire s’effectue avant tout sur sa region posterieure (zone ano-ge£nitale, queue, croupe: 45 % des observations), anterieure (garrot, face, front ou oreilles: 38 %), mais rarement sur les pattes ou le milieu du corps. Quelques lechages entre femelles font partie du comportement agonistique et menent directement A la chasse ou & la morsure du partenaire; dans ce cas, ıls sont effectues de preference sur la region posterieure. La totalıte des lechages des regions anterieures du partenaire et une partie des autres lechages representent l’allogrooming. Ce comportement est donc peu frequent chez cette espece, puisqu’il n’a et€ observ& qu’ä 22 reprises; ıl est &galement de courte duree et ıl n’est jamais reciproque. Le marquage du partenaire est tout & fait exceptionnel (5 observations) et n’a Et€ observe que chez les femelles. Un individu peut frotter sa glande preorbitaire contre le tarse ou la croupe d’une autre (4 cas); ou bien elle frotte son front contre le corps du mäle (1 cas). Il n’y a donc aucun marquage d’une femelle par le mäle, nı d’un jeune par un adulte. Contrairement ä d’autres especes de Cephalophes, il n’y a jamais de marquage reciproque. 36 G. Dubost et F. Feer Comportement sexnel Le mäle adulte pratique un contröle plus ou moins permanent des differentes femelles en les sentant ou les lechant ä la croupe. Il effectue parfois une breve poursuite sexuelle de l’une d’entre elles, m&me sı elle n’est pas r&ceptive. Il s’approche alors par l’arriere, en attitude Etiree, le museau et le cou tenus horizontaux A un niveau plus bas que les Epaules, comme le sıgnale Dittrich (1972). Il effectue de nombreux coups de langue ä vide (Fig. 3). La temelle fuit en soufflant, la tete basse. Lorsque le mäle parvient ä la rejoindre, il tente de la lecher dans la region genitale ou sur les cuisses et ıl effectue m&me quelques essaıs de monte. Ce comportement ne dure generalement guere plus de quelques minutes. Fig. 3. Poursuite sexuelle. Le mäle effectue un lechage & vide (dessin d’apres photo) Les m&mes Elements fondamentaux de cour se rencontrent lorsque la femelle est proche de l’oestrus, mais ıls sont alors beaucoup plus frequents, les lechages du mäle s’etendant sur toute la region posterieure de la femelle: tarses, zone mammaire. La poursuite est alors soutenue et vive, et se faıt parfois A la course. La femelle fuit tete basse en souftlant ou gemissant. Le mäle souffle ou gemit egalement, la bouche entrouverte; il se leche souvent la face et les glandes preorbitaires avec de grands coups de langue. La poursuite se fait ainsi en ligne droite ou en arc de cercle, selon des deplacements «en accordeon» du couple, le mäle s’approchant lentement de la femelle en position &tiree pour lui lecher l’arriere-train, la femelle fuyant alors vivement. Elle peut durer sans interruption une h ou plus avant qu’intervienne une pause au cours de laquelle le mäle reste trequemment fıge aupres de la femelle, tete et cou horizontaux pres de sa vulve, parfois en Erection et poussant de petits cris. Apres un bref moment d’ımmobilite, ıl peut se rapprocher brusquement d’elle en faısant 1 ou 2 petits pas ou pietinements sur place avec ses pattes anterieures. Si la femelle s’est couchee, le mäle reste debout ä ses cötes ou se couche lui-m&me pour peu de temps avant de la faire se lever et repartir. D’autres pauses sont directement provoque&es par la miction de la femelle qui induit ainsı regulierement l’arret et le Flehmen du mäle. Sı la femelle a reussi a &chapper au mäle, ce dernier la retrouve rapıdement ä la trace. L’intensite du comportement du couple attire frequemment les autres anımaux: jeune venant jouer autour des adultes, fe- melle adulte venant attaquer celle qui est poursuivie par le mäle ou, au contraire, cette derniere agresse et monte l’arrıvante qui peut riposter en tournant sur elle- m&me et en prenant la position de defense tete basse. DiTTRicH (1972) note que la femelle en oestrus peut egalement monter le mäle, de möme quelle peut le sentir dans la region du p&nis ou effectuer quelques mouvements agressifs A son €gard (coups de front dans les flancs). Mais, contraire- Fıg. 4. Monte (dessin d’apres photo) ment aux observations de cet auteur de Comportement de C. rufılatus et varıabılıte du genre Cephalophus 97 coups de tete ou de cornes de la part du mäle sur la femelle, nous n’avons jamais observe de gestes agressifs de la part du mäle. La copulation peut avoir lieu lorsque la femelle ne fuit plus. Avant de monter, le mäle se tiend en Erection, la queue Ecartee en crochet du corps; ıl leche le dessus de la croupe et la racine de la queue de la femelle. La monte est breve, tete dress&e, museau pointe vers le dos de la femelle. A l’&jaculation, le mäle se redresse plus ou moins & la verticale (Fig. 4); apres la demonte, il se leche le penis. Plusieurs montes se suivent ä bref intervalle, precedees chaque fois par un lechage du dessus de la croupe et de la queue de la femelle. La receptivite de la femelle dure 2 jours environ (et non 0,5 a 1 journee, comme le signale DiTTricH 1972). Durant toute cette periode, la femelle montre un comportement tres different de la normale (Tableau 1). Elle n’est pas plus active que d’habitude, mais elle marque un peu moins et urine, par contre, bien davantage. Elle approche beaucoup moins les autres anımaux et est moıins souvent approchee par eux (sauf par le mäle). Elle les löche rarement dans la region anterieure ou moyenne du corps, mais plus souvent au pöle anal; elle n’est plus lechee que par son mäle. Enfin, elle est beaucoup moıns agressee par eux et les agresse, par contre, bien plus en donnant beaucoup de coups de tete et en effectuant de nombreuses montes. Tableau 1. Frequence relative des differents comportements d’une femelle avant et pendant sa periode de receptivite au mäle (par heure d’activite) Comportement non receptive debut receptivite oestrus % activite 46,3 Mictions 1,1 Marquages pr£orbitaires 46,5 Approche 1 femelle 1 mäle Est approchee par 1 femelle par 1 mäle Hume 1 femelle 1 mäle Est humee par 1 femelle par 1 mäle Est lechee par 1 femelle par 1 mäle Agresse 1 femelle 1 mäle Est agressee par 1 femelle 1,6 0 0,8 tres souvent 0,8 tres souvent 0 tres souvent 0,7 tres souvent 0 tres souvent 0 tres souvent 0,5 0,2 0,2 Comportement agonıstique A l’exception des gestes agressifs plus ou moins ritualises de la femelle en oestrus envers le mäle, les rapports entre individus de sexe different sont presque toujours pacifiques. Cependant, le mäle adulte peut harceler particulierement un individu (par exemple la jeune femelle F4, ägee de 10 mois Y2), sans que l’on en comprenne les raisons. 38 G. Dubost et F. Feer Entre femelles, par contre, les explications A tendance agonistique sont habituelles. On peut distinguer par ordre croissant d’agressivite: — le contröle et le lechage de la croupe annongant une morsure ou une chasse. — la monte. — le lever: un anımal couch& est oblige de se lever, pour ceder sa place ou fuir, en &tant simplement approche par un autre, hum& ou touche ä& la croupe. — la chasse: un individu poursuit un autre anımal au pas ou ä la course. Le poursuivi fuit et cherche & se cacher. — le coup de tete: donne avec le front dans le garrot, le flanc, la Fig. 5. Coup de tete d’une femelle contre l’&paule d’une autre croupe ou entre les pattes ar- (dessin d’apres photo) rieres (Fig. 5). lesmalespeut ainsı frapper de ses cornes. — Ja morsure: dans la region anale, sur la croupe, les pattes arrıeres ou la queue, parfois egalement sur les oreilles. Tous ces comportements, exception faite du lechage et des montes agressives absentes chez le mäle, sont presents dans les deux sexes. Les chasses sont le comportement le plus frequent (38 A 71 % des interactions agonistiques chez les femelles, 63 % chez le mäle); les coups de tete et morsures viennent ensuite (respectivement 27 et 13 % chez les femelles, 16 % chez le mäle). Le lever hierarchique est rare (3 10 % chez les femelles, 5 % chez le mäle), de m&me que la monte (3 9 % chez les femelles). Ainsı les comportements agonistiques les plus engages et les plus violents sont generalement les plus frequents, ce qui est souvent les cas chez les especes solitaires qui, contrairement aux especes sociales, ont rarement rıtualise leur comportement agressif. Chez C. rufılatus, ıl n’y a pas d’attitude demonstrative de menace ou de parade, exception faite de la position defensive decrite plus haut, le front vertical, nez au sol. Un individu est donc directement attaqu& par un autre et le seul premice au comportement agressit de ce dernier est son approche. Sı l’ındividu menace& est sup£rieur, il ne fuit pas mais menace, tete basse. L’anımal domine cede le terrain, cherche ä s’Echapper et & se cacher, se couche sur le sol, tete et cou Etendus & plat devant luı. Iln’y a jamais eu de combat proprement dit au sein de cet Elevage, mais une explication assez longue entre 2 femelles avait degenere en un bref combat frontal, les 2 aniımaux tournant sur eux-m&mes en cherchant & se frapper sur le corps. Compte-tenu des comportements agressıfs observ&s, on peut £tablir l’ordre hierarchi- que suivant: mäle Mi > femelle F3 > femelle Fi > femelle F2. La femelle F2 effectue quelques comportements agressifs sur la femelle Fl, mais 4 foıs moins que l’inverse; la jeune femelle F4 est hors de la hierarchie femelle et n’est agressee que par le mäle MI. Vie sociale Nature et importance des contacts sociaux En captivite, l’exiguite du terrain contraint les anımaux & des contacts auxquels ıls echapperaient dans des conditions naturelles. C’est sans doute la raison pour laquelle nous avons observ& beaucoup de comportements agressıfs, meme entre femelles apparentees, ce qui laisse supposer que les femelles filles doivent quitter le domaine ou territoire de leur m£re, sans doute au plus tard une fois qu’elles sont devenues adultes. Ceci est encore plus Comportement de C. rufılatus et varıabilite du genre Cephalophus 99 vrai pour les mäles, puisque les jeunes mäles comme M2 doivent £tre separes de leur pere. Malgre ces inconv£nients, l’observation des anımaux libres sur un tel terrain permet de percevoir quelques unes des caract£ristiques sociales de l’espece, principalement en ce qui concerne les possibilites de vie en groupe et le choix des partenaires. La majorite des contacts sociaux entre individus sont des rapprochements simples, oü un d’entre eux se dirige vers un autre, passe ou s’arrete aupres de luı (en moyenne 4,1 rapprochements par anımal et par heure d’activite). Les contröles olfactifs sont bien moins frequents (1,0/animal/heure), suivis par les lechages (0,5/anımal/heure) et enfin les com- portements agressifs (0,4/anımal/heure). Les contacts sociaux sont donc d’autant plus frequents qu’ils reclament peu d’engagement physique. En prenant en compte toutes les interactions sociales se produisant entre individus, on peut £tablır une sorte de hierarchie de l’activite sociale des individus, selon la frequence des rapports que chacun d’entre eux a avec ses partenaires (par heure d’activite). La femelle la plus ägee, Fl, est la plus active (16,2 interactions en moyenne par heure d’activite), et ce niveau d’activite sociale decroit regulierement avec l’äge chez les femelles: F2 14,3 interactions en moyenne; F3 13,0; F4 2,24. Le mäle adulte MI se situe A un nıveau moyen de 13,7. Les differentes activites d’un individu donne sont le plus souvent independantes de celles des autres. Il est significatif A ce point de vue qu’il n’existe pour ainsı dire pas d’allomımetisme des activites, exception faite lorsque 2 anımaux se trouvent directement engages dans un comportement social precis (agonistique ou sexuel), auquel cas les 2 partenaires effectuent les m&mes comportements A peu pres au meme moment et au meme endroit: marquages, miction, defecation. De m&me, la rarete des lechages et des marquages d’un autre anımal, le caractere le plus souvent unilateral de ces comportements, la nature essentiellement solitaire des jeux qui ne deviennent jamais contagieux, montrent que chaque individu mene une vie avant tout solitaire, ce qui correspond aux indications que l’on a sur son mode de vie dans la nature. Cependant, plusieurs comportements indiquent l’existence d’une certaine liaison sociale. Durant la journee, sı les individus peuvent prendre des repos ısoles en plein air, ıls se couchent souvent @galement dans les abris. Dans ces derniers, ıls se reposent en moyenne aussi souvent seuls qu’en compagnie d’un ou plusieurs autres, et la proportion de repos isoles ne varıe pas significativement d’un anımal & l’autre (40,0 aA 66,7 %, N = 287), si ce n’est que les tout jeunes se reposent plus souvent que les adultes dans un gite particulier. Il n’y a done nı recherche nı Evitement des partenaires pour le repos de jour. Les quelques observations effectu&es la nuit par contre, c’est A dire lorsque le nıveau general d’activite est reduit ou nul, indiquent que le regroupement des anımaux pour le repos est plus fort, puisqu’on rencontre souvent jusque 5 indıvidus couches dans le m&me gite. Bien qu’ils circulent generalement seuls, les individus sont appeles ä se rencontrer frequemment au cours de leurs actıvites. Beaucoup de rapprochements sont le fait delibere d’un animal. En dehors des periodes speciales comme l’oestrus d’une femelle ou la frequence des rapprochements entre cette femelle et les autres anımaux est tout & fait particuliere (cf. Tableau 1), les differents individus effectuent globalement le m&me nom- bre de rapprochements des autres anımaux, exception faite des tout jeunes qui restent le plus souvent couches a l’Ecart. Chaque anımal est a peu pres aussi souvent approch& par ses semblables qu’il ne les approche lui-m&me. La seule difference individuelle se remarque lorsqu’une femelle vient de mettre bas (F1 en septembre, F2 en juin). Dans ce cas, la me£re s’approche peu des autres anımaux mais elle est, par contre, plus souvent approch&e par eux. Le rapport Nombre de rapprochements subis/Nombre de rapprochements inities passe ainsı de 1,00 & 1,80 chez Fl, et de 0,51 a 2,67 chez F2 entre l’Epoque oü la femelle etait seule ou gestante et celle ou elle possede un jeune. Cela doit correspondre ä& la tendance naturelle l’isolement des me£res, comme cela existe chez de nombreux ruminants et a Et€ constate chez C. monticola (DusosT 1980). Il en est a peu pres de m&me pour les contröles olfactifs d’un autre anımal. Mais, si l’on 40 G. Dubost et F. Feer constate la m&me inversion du rapport des contröles olfactifs recus par rapport ä ceux qui sont eftectues chez les femelles nouvellement meres compar&es aux autres anımaux (rapport passant aınsi de 1,16 a 2,46 chez Fl, et de 0,51 a 3,49 chez F2), les contröles olfactifs se distinguent des rapprochements en ce que les mäles effectuent toujours beaucoup plus de contröles des autres anımaux qu’ils en sont l’objet de leur part (4,8 a 5,6 foıs plus). Les contröles oltactifs sont plus souvent faits par les adultes que par les jeunes, et en particulier par le mäle adulte. Parmi les femelles, le nombre de contröles effectues croit avec l’äge. Les individues de rang social superieur contrölent avant tout leurs partenaires dans la region posterieure du corps (67 % des contacts; N = 164), ce qui correspond peut- etre A une persistance des rapports mere-jeune, et ce qui rappelle beaucoup les rapports mäle-femelle; au contraire, 56,1 % des contröles effectues par des inferieurs sur leurs sup£rieurs se font dans la region c&phalique (N = 82). Lorsque le contact socıal est un peu plus engag& comme l’est le lechage d’un partenaire, 3 categories sociales bien distinctes se degagent: femelles, jeunes et mäle adulte. Les femelles font beaucoup plus de lechage social que les mäles, et chez elles la frequence de ce comportement augmente regulierement avec l’äge, comme pour les contröles olfactifs. Globalement, chaque femelle etfectue autant de lechages des partenaires qu’elle n’est elle- me&me lechee par eux, mais ıl faut rappeler que les lechages sociaux ne sont jamais reciproques. Les jeunes anımaux lechent tres rarement les adultes, mais ıls sont abondam- ment leches par eux. Au contraire, le mäle adulte effectue de nombreux lechages des femelles, m&me en dehors des poursuites sexuelles, mais ıl n’est jamaıs leche. Le lechage d’un jeune par un adulte quelconque se fait presque exclusivement dans la region posterieure (24 observations sur 28). Comme pour les contröles olfactifs parmı les adultes et subadultes, les anımaux superieurs lechent surtout l’arriere-train de leurs inferieurs (52,2 % des cas; N = 23), les inferieurs lechant avant tout la partie anterieure du corps de leurs superieurs (60 %). Les marquages d’un partenaire au moyen des glandes preorbitaires n’ont et observes qu’entre 2 femelles adultes, la fille F2 ayant marque sa mere Fl A 3 reprises, cette derniere une seule fois sa fille. Comme pour le lechage, on reconnait 3 types fondamentaux de rapports sociaux entre les individus, lorsqu’on considere les comportements agonistiques. Les jeunes ne sont pas agressifs et le mäle adulte n’est jamais agresse. La femelle la plus ägee F1 s’est toujours montr&e comme la plus agressive, suivie par la jeune femelle F3; par contre, F2 &tait de loın la plus agressee. Globalement, les 2 femelles les plus ägees (Fl et F2), c’est & dire les 2 femelles ayant des jeunes, sont les indıvidus les plus souvent impliques dans des explications agonistiques avec un autre anımal, mais surtout entre elles-m&mes. Il n’appa- rait pas, cependant, qu’une femelle nouvellement me£re connaisse un surcroit d’agressivite A l’egard de ses semblables, puisque le comportement de ces 2 femelles n’a jamais varıe en consequence. Preferences sociales L’observation du comportement des differents individus de cet Elevage, qui se trouvaient etre tous directement apparentes les uns aux autres comme le sont sans doute la plupart des unites sociales dans la nature, nous permet de mettre en Evidence certaines preferences (Tableau 2). La femelle F1, la femelle la plus ägee et, directement ou indirectement, la fondatrice de ce groupe d’animaux, effectue une grande partie de ses lEchages sur sa fille F4 encore immature. Elle semble cependant marquer une nette preference sociale pour sa fılle adulte F2, aupres de laquelle elle se repose souvent (comme d’ailleurs aupres du mäle adulte M1); elle la recherche, la contröle et la leche plus souvent que les autres. Cependant, les rapports qu’elle entretient avec elle sont surtout de nature agressive. La femelle adulte F2 leche beaucoup sa jeune «soeur» F4. Elle se repose souvent en compagnie de sa fille subadulte F3 qu’elle recherche et leche plus frequemment que les Comportement de C. rufılatus et variabihite du genre Cephalophus 41 Tableau 2. Preferences sociales: % de comportements adresses par un animal a un autre Receveur Comportement Inıtiateur Femelle FI FemelleF2 FemelleF3_ FemelleF4 Mäle Mi IE Repos 16,0 31,4 Rapprochement 43,1 41,4 Flairer 32,4 3159 Lecher ; 0 81,3 Agresser 5,0 Repos 28,6 Rapprochement il Flairer 36,2 Lecher Agresser Repos Rapprochement Flairer Lecher Agresser Repos Rapprochement Flairer Lecher Agresser Repos Rapprochement Flairer Lecher Agresser autres anımaux. Ses rapports envers sa mere Fl sont pacifiques (contröles olfactifs) ou agressifs. La femelle subadulte F3 se montre plus attiree par les 2 femelles adultes precedentes que par le mäle adulte ou la jeune femelle F4. Elle se repose frequemment aupres de sa mere F2, mais recherche beaucoup sa grand-mere F1 lors des activites. Avec ces 2 femelles adultes, ses rapports sont A la fois pacifiques (contröles oltactifs, lechages) et agressifs. La jeune femelle F4 se repose surtout en compagnie de F3, c’est ä dire l’autre jeune femelle non suitee, mais elle contröle frequemment tous les adultes. Quant au mäle adulte, il ne montre pas de preference nette pour un anımal donne ni pour ses repos, nı pour ses contröles olfactifs, nı pour ses contacts pacifiques ou agonistiques. Cela laisse presager que cette espece ne vit pas en couple et que, dans la nature, un mäle aurait un domaine recouvrant ceux de plusieurs femelles. Cependant, a l’egal d’autres especes solitaires comme Muntiacus muntjak et M. reevesi (Dusost 1970, 1971) et peut-Etre Aydropotes inermis (FEER 1982), on remarque chez lui, au nıveau des rapprochements et des lechages, une certaine preference pour la femelle Fl, c’est & dire la plus ägee et celle qui a la plus forte descendance. Faute d’observations effectuees en liberte, nous ne savons pas si cette m&me preference existe dans des conditions de vie naturelle et sı elle represente un rudiment de vie de couple. Cependant, la temelle C. rufilatus est capable de connaitre un nouvel oestrus apres la mise-bas (4,5 semaines selon DITTRICH 1972; mais 6 jours seulement selon nos propres observations). Cela explique peut-Etre pourquoi le mäle M1 a montre une forte preference pour la femelle F1 en septembre 1982, 7 semaines apres que cette femelle ait mis bas son jeune F4, alors qu’en juin 1983, la femelle F2, qui avait mis bas 8 jours auparavant, se trouvait A son tour la principale cıble des rapproche- ments de la part du mäle. 42 G. Dubost et F. Feer Les rapports existant entre les 3 ftemelles F1, F2 et F3 denotent ä la fois une recherche pacıfique du partenaire et un comportement agressıf d’eloignement. Les 2 meres Fl et F2 ont des rapports plus souvent agressıfs que pacifiques: en moyenne, 74,7 % des comporte- ments agressifs ou menant A une agression effectues par la femelle F1 envers les autres anımaux ont etE dıriges sur la femelle F2 contre 28,3 % seulement de ses comportements pacifiques, et 71,8 % des comportements agressifs de F2 ont dt adresses A Fl contre 19,6 % des comportements pacıfiques. Vis ä vis de la femelle subadulte F3, les rapports des 2 temelles adultes F1 et F2 sont, au contraire, plus pacifiques qu’agressifs: 17,1 % des comportements pacifiques contre 8,9 % des comportements agressifs de F1 se sont portes sur F3; 30 % des comportements pacifiques contre 12,8 % des comportements agressifs de F2 ont et€ adresses A F3. Le m&me phenomene, encore plus accuse, est valable pour les rapports de Fl ou F2 avec la jeune femelle F4: 35,7 % contre 9 % pour Fi vis ä vis de F4, et 30,9 % contre 5,2 % pour F2 vis & vis de FA. Par contre, la femelle subadulte F3 entretient avec les femelles adultes FI ou F2 des rapports d&ja plus agressifs que pacifiques: 44 % contre 33 % vis A vis de Fl, et 43 % contre 32 % vis a vis de F2. Il apparait donc que l’importance des rapports agressifs entre femelles croit avec l’äge de la fille et est surtout le propre de cette derniere, puisque les adultes se montrent toujours beaucoup plus pacifiques envers elle. Il se pourrait donc que, dans la nature, ce comportement engendre l’ısolement spatial des m£res et filles, une fois ces dernieres devenues adultes. Discussion Il est couramment admis que les Cephalophes et Neotragines afrıcains repre&sentent les equivalents des Cervides forestiers d’Asie ou d’Amerique du sud. En fait, cette ressem- blance est, le plus souvent, beaucoup plus apparente que reelle et ne semble &tre que le resultat des fortes contraintes physiques du milieu qui modelent oblıgatoirement les organısmes. En pratique, le groupe des Cephalophes afrıcains est bien plus riche que les autres groupes @quivalents des autres continents, puisqu’il comprend souvent 4-6 especes sympatriques bien separ&es Ecologiquement les unes des autres, 1 ou 2 d’entre elles etant me&me plus ou moins sp&cıalisee a un biotope particulier, alors que les Cervides de m&me poids n’ont qu’l ou 2 formes & Ecologie assez generale. Les connaissances des Cervides sud-americains ne sont pas suffisantes pour que l’on puisse Etablir une comparaison entre eux et les Cephalophes, et ıl en est de m&me de la plupart des especes asiatiques. Mais, si l’on consid£ere le genre Muntiacus, et principalement M. muntjak qui peuple les forets indo- malaises (BARRETTE 1977; van BEMMEL 1952; BLANFORD 1888-91; DuBosTt 1971; LyDEk- KER 1924; PRATER 1965), on constate que C. rufllatus est le Cephalophe qui se rapproche le plus de cette espece par: — sa taille: 10-14 kg contre 15-35 kg - sa coloration: corps rougeätre, front et pattes gris-noirätres chez les 2 especes — ses appendices frontaux: presents seulement chez les mäles, reduits a de petits boutons osseux chez les femelles - son rythme d’activite: A predominance diurne, mais egalement actıf de nuıit — ses biotopes: de la grande for£t & la fort claire, avec une utilisation frequente des zones decouvertes - son regime alımentaire mixte: fruits, feuilles, parfois herbe — son comportement agonistique: coups de t&te et morsure chez les mäles comme chez les femelles - son mode de vie sociale: surtout solitaire, mais regroupement possible de plusieurs individus. En captivite, forte intolerance des mäles, cohabitation des femelles avec agression caracterisee entre certaines. L’&quivalence physique et Eco6thologique entre ces 2 especes semble donc assez grande, Comportement de C. rufilatus et varıabılite du genre Cephalophus 43 bien qu’il persiste certaines divergences, surtout en ce qui concerne le comportement social: mode de marquage du domaine, importance de l’allogrooming reciproque et du marquage des partenaires, M. muntjak se montrant nettement plus riche dans ses relations interindividuelles. Cependant, dans bien des comportements, C. lan est tout & fait semblable aux autres Cephalopes d&jä etudies comme C. monticola et C. dorsalis (Dusost 1983), et cela atteste de la grande homog£neite du genre Cephalophus, au moins en ce qui concerne les comportements de base (attitudes corporelles). Chez ces 3 especes de Cephalophes, comme chez toutes celles observ&es, on retrouve, en effet, le m&me schema fondamental lors du coucher: simple, sans cer&monial annexe (Egalement chez C. callıpygus et nigrifrons, obs. pers.); — lors de la recherche de la nourriture: parfois avec grattage du sol ou lever sur les pattes arrieres (egalement chez C. maxwelh, Hopkıns 1966; C. callipygus et nigrifrons, obs. pers., Kınapon 1982); - lors de l’alarme: corps bloque & l’arret, une patte anterieure levee (WALTHER 1979; Egalement chez C. callipygus et nigrifrons, obs. pers.); - lors de la miction et de la defecation: attitudes engag&es pour la miction mäle ou femelle, mais non pour la defecation, sans comportement annexe et sans dissemination des laissees (Egalement chez C. callıpygus et nıgrifrons, obs. pers.; maxwelliı, AESCHLIMANN 1963; harveyı, Kıncpon 1982); - lors du Flehmen: bref, bouche peu entrouverte, present seulement chez le mäle, le plus souvent unique (Egalement chez C. maxwelli, obs. pers., AESCHLIMANN 1963); — lors du marquage pr£&orbitaire: frottement contre un support vertical, disperse partout, present dans les 2 sexes, le plus souvent unique (Egalement chez C. callipygus et nigrifrons, obs. pers.; maxwelli, AESCHLIMANN 1963; Rarıs 1974); — lors de la poursuite sexuelle: en ligne ou en cercle, lechage abondant de la region posterieure de la femelle, position etiree du mäle en Überstrecken, soufflements des 2 partenaires, comportement agressif plus ou moins ritualise de la femelle sur le mäle (egalement chez C. zebra, SCHWEERS 1984), copulation breve avec la tete dressee (Egalement chez C. maxwelli, obs. pers.; nigrifrons, WALTHER 1979); — lors des rapports agonistiques: absence d’attitudes de menace ou de parade (Egalement chez C. maxwelli, Raııs 1975), posture defensive nez au sol, front vertical (Egalement chez C. nigrifrons, WALTHER 1979), soumission tete et corps aplatıs sur le sol. Cependant, entre ces 3 especes comme entre d’autres, apparaissent certaines differences qui sont le reflet des divergences dans les modes de vie (biotopes, rythmes d’activite), et qui s’inscrivent en majorite dans le comportement socıal (Tableau 3). Dans la plupart des cas, C. rufilatus est place a mi-chemin entre l’espece diurne C. monticola et l’espece nocturne C. dorsalıs: sur 36 caracteres analyses, 16 sont directement interme&diaires entre ceux de ces 2 especes, 6 correspondent davantage A ceux de la forme diurne et 8 äA ceux de la forme nocturne. Il apparait ainsı que ces modifications comportementales semblent la conse- quence directe du changement de rythme d’activite (Dusost 1983). Sur 6 caracteres seulement, C. rufilatus se distingue nettement de C. montıicola et dorsalıs: absence de jeux sociaux, absence d’allogrooming reciproque, nombreux comportements agonistiques entre anımaux apparentes, presence de morsure, monte des autres anımaux par une femelle en oestrus, defecation localisee. Exception faite de la morsure qui existe aussı chez C. zebra et nigrifrons ( WALTHER 1979; SCHWEERS 1984), tous ces comportements apparaissent pour l’instant comme caract£erıstiques de l’espece rufılatus. Deux d’entre eux doivent £&tre particulierement notes. La relative bonne entente des femelles en captıvite joint au fait que, contrairement A toutes les autres especes congeneriques, les femelles sont totalement inermes, peut nous faire supposer que l’espece pourrait ne pas montrer de territoire femelle, contrairement ä ce qui est connu chez C. monticola et ce qui est peut-£Etre le cas de C. maxwelli (AEscHLI- MANN 1963), callipygus et dorsalıs (obs. pers.). C. rufılatus utilise volontiers en captıvite, comme dans la nature selon les quelques observations disponibles, les surfaces plus ou moins decouvertes; ıl mange de l’herbe et est 44 G. Dubost et F. Feer Tableau 3. Variabilite comportementale entre differentes especes de Cephalophus C. monticola C. rufilatus C. dorsalıs foret om- brophile Milieux Caracteristiques physiques Poids corporel Presence de cornes 4-6 kg -? Activite Rythme diurne % de temps actıf lieu de repos Attitudes attıtudes corporelles accusees repos sur les carpes - deplacements tete hori- zontale Informations visuelles Reconnaissance iA distance Signal corporel optique Marquage avec les cornes Informations olfactives Hume l’air Defecation en tas Marquage pre£orbitaire Informations acoustiques Cri de contact Soufflement d’inquietude Sifflement d’alarme Tapement du sol avec les pattes Jeux Elements Course, saut, rua- de, mar- quage, monte, combat foret clai- re, galerie forestiere, bordure de foret 10-14 kg 6) diurne- nocturne? 45-70 % a decou- vert ou cache simples tete hori- zontale moyenne + 8-2? repete lors des rap- proche- ments + Course, saut, marquage, combat foret om- brophile 19-25 kg Ö=2 nocturne 33-54 % le plus souvent cache simples + tete hori- zontale ou basse mauvaise unique lors des rap- proche- ments Rare Rare Course, saut, combat autres especes 8-2 diurnes: callipygus, lencogaster, nigrifrons (3); adersi, natalensıs (5) diurne-nocturnes: maxwell (1), sylvicultor (3), harveyi (5) nocturnes: nıger (2), spadıx (5) -: callipygus, +: nigrifrons (3) tete horizontale: callıpygus, maxwelli, nigrifrons (3) tete horizontale ou basse: sylvicultor (3) bonne: callıpygus, nigrifrons (3) mauvaise: sylvicultor (3) — chez les autres especes, sauf maxwell (1) S maxwellh, 8-2 callipygus, ? nigrifrons (3) +: nigrifrons (3) =: callıpygus, nigrifrons (3) 3-2 maxwelli (1,7); 8-2 cal- hpygus, 2 nigrıfrons (3) +: callıpygus (3) +: maxwelli (1); callıpygus (3); nigrifrons, sylvicultor (5) +: callıpygus (3) Comportement de C. rufilatus et varıabılite du genre Cephalophus 45 Tableau 3 (countinued) C. monticola C. rufilatus C. dorsalıs autres especes Frequence + _ _ Participants Jeuneset Jeunes Jeunes adultes Jeux sociaux + _ Rares +: maxwell (1); callıpygus, nigrifrons (3) Contacts avec 1 partenaire Allogrooming reciproque +: maxwelh (1,7); callipygus, nigrifrons (3) Marquage du partenaire : callipygus (3); zebra (9) Marquage du jeune : max well (1); +/-: zebra (4,9) Marquage reciproque : maxwellı (1,7), zebra (9) Poursuite sexuelle Laufschlag Ebauche : zebra (9); nigrifrons (10, 11) Pose menton/croupe - : sylvicultor, zebra (9) Remontee laterale du mäle _ : zebra (97 Comportement agressıf _ rufılatus (2), dorsalıs (4), du mäle sur femelle maxwelli (6), zebra (9) Monte des autres anımaux par la femelle Arret de la femelle pour la copulation Comportement agonıstique Marquage mutuel ? +: maxwell (9) Morsure 2? (ritua- +: d zebra (9); 2 nigrifrons (11) lise) Coups de front ou cornes +? +: d-2 maxwelh (8); d zebra (9) Vie sociale Goupement en nature 1 (2-3?) 14: maxwelli (1); natalensıs, sylvicultor (5) Groupement en captivite + +/- = Allomimetisme + Rare _ % derepos du mäleavece 75% 42 % 20% les femelles (1): AESCHLIMANN 1963; (2): DITTRICH 1972; (3): obs. pers.; (4): FRÄDRICH 1964; (5): KINGDON 1982; (6): Rarıs 1970; (7): Rarıs 1974; (8): Rarıs 1975; (9): SCHWEERS 1984; (10): WALTHER 1968; (11): WALTHER 1979 attire par les pierres ä sel. Ceci le differencie beaucoup des especes forestieres proprement dites que nous avons observe&es (C. monticola, nigrifrons, dorsalıs et callipygus). C’est donc une espece beaucoup plus savanıcole que les autres; et ıl est a prevoir que son comporte- ment consistant A concentrer fortement ses laissees sur de petits emplacements, comme il apparait en captivite, doit egalement se rencontrer dans la nature, puisqu’un tel mode de depöt des feces est presque toujours la caracteristique d’anımaux de regions ouvertes. Globalement, le genre Cephalophus se montre donc & la foıs tres homogene dans ses comportements fondamentaux plus ou moins fixes comme les modes de locomotion, d’excretion, de marquage, de reproduction, etc., et tres variable dans les comportements qui subissent une certaine varıation selon les conditions de vie. Cela corrobore precisement les resultats obtenus ä partir des caracteristiques physiques, puisqu’on rencontre toujours chez eux, ä cötE du m&me schema de base, de nombreuses modifications selon les preferences Ecologiques ou le rythme d’activite. Ainsı peuvent coexister des caracteres que l’on peut interpreter comme «archaiques» ou, au contraire, «&volues» (DuBost 1983). Mais 46 G. Dubost et F. Feer cette variabilit€ au sein du m&me genre nous oblige a reconsid£rer la valeur phylogenetique et Evolutive de certains comportements, puisque nous n’avons aucun argument pour decider si, A l’interieur de ce genre, une ou plusieurs especes sont plus avancees que d’autres. Au vu de la dualite d’expression qui peut se manifester lors du comportement sexuel (presence ou non de Laufschlag ou de pose du menton sur la croupe), du comportement agonistique (coups de front seuls ou presence simultanee de coups de front et de morsures), ou du comportement social (vie solitaire ou vie plus ou moins permanente avec 1 ou plusieurs partenaires, lechages et marquages mutuels absents ou frequents, allomimetisme ou non dans les actıvites, etc... .), il semble bien que le genre Cephalophus se trouve place A un carrefour Evolutif. Chez lui, des comportements tres «avances» peuvent ainsi apparaitre selon les especes et leurs conditions de vie, et ıl ne semble plus possible de considerer les Cephalophes comme les Bovides les plus primitifs. Remerciements Nous remercions vivement Mr et Mme SAINnT-PıE pour leur accueil et les facilites qu’ils nous ont offertes lors de nos sejours au Zoo de Pietat. Zusammenfassung Verhaltensunterschiede innerhalb der Gattung Cephalophus (Ruminantia, Bovidae) am Beispiel von C. rufilatus Gray, 1846 Im Zoo von Pietat konnte eine auf einem 0,7 ha großen Gelände freilebende Gruppe von 7 Cephalophus rufilatus ethologisch untersucht werden. In den meisten seiner Verhaltensweisen unter- scheidet sich diese Art nicht von den anderen bereits untersuchten Duckern. Jedoch zum Unterschied von eben diesen Arten zeigt rufilatus ein fast völliges Fehlen sozialen Spieles und des sich gegenseiti- gen Putzens. Vorhanden sind dagegen Beißen und Kopfstöße im Verlauf agonistischer Auseinander- setzungen. Der Laufschlag ist nur rudimentär vorhanden. Das in Brunft befindliche Weibchen besteigt häufig die Gruppenmitglieder. Die Defäkation erfolgt stark lokalisiert in Häufchen. Global betrachtet ist rufllatus eine Art mit Tag-Nacht-Rhythmus und zeigt dementsprechend auch ein intermediäres Verhalten. Dies zeigt sich besonders im Sozialverhalten, denn die einzelnen Individuen vertragen sich relativ gut, wenngleich auch jedes Tier seine eigene von den anderen unabhängige Aktivität hat und mit diesen nur ein Minimum inter-individuelle Kontakte hat. Es scheint keine bestimmte Wahl des Partners zu bestehen. Im großen und ganzen erscheint die Gattung Cephalophus relativ homogen in bezug auf funda- mentale Verhaltensweisen wie Fortbewegung, Sekretausscheidung, Markierung, Fortpflanzung, usw. Unterschiede dagegen findet man bei allen den Lebensstil betreffenden Verhaltensweisen im Biotop und bei den Aktivitätsrhythmen. Dies alles entspricht recht gut der morphologischen und ökologi- schen Variabilität dieser Gruppe. Neben Merkmalen, die man als «ursprünglich» bezeichnen kann, findet man auch gewisse höher entwickelte Verhaltensweisen. In diesem Kontext erscheinen die Cephalophinae innerhalb der Bovidae keineswegs als eine der primitivsten Gruppen. Bibliographie AESCHLIMANN, A. (1963): Observations sur Philantomba maxwelli (Hamilton-Smith), une Antilope de la for&t eburneenne. Acta tropica 20, 341-368. BARRETTE, C. (1977): Some aspects of the behaviour of muntjacs in Wilpattu National Park. Mammalıa 41, 1-34. Baupenon, P. (1958): Ecologie des petites et moyennes Antilopes dans le Togo meridional. Mammalıa 22, 285-293. BEMMEL, A. van (1952): Contribution to the knowledge of the genera Muntiacns and Arctogalıdıa in the Indo-Australian Archipelago (Mammalia, Cervidae and Viverridae). Beaufortia 16, 1-50. Brancou, L. (1958): Distribution geographique des Ongules d’Afrique Equatoriale frangaise en relation avec leur eEcologie. Mammalıa 22, 294-316. Brancou, L. (1962): Mensurations et poids de quelques Ongules du tchad et de la Republique Centre-africaine. Mammalıa 26, 84-106. BLANFORD, W. T. (1888-1891): Mammalia. In: The fauna of British India, including Ceylon and Burma. London: Taylor and Francıs. Dittrich, L. (1972): Beobachtungen bei der Haltung von Cephalophus-Arten sowie zur Fortpflan- Comportement de C. rufilatus et variabılite du genre Cephalophus zung und Jugendentwicklung von C. dorsalis und C. rufilatus in Gefangenschaft. Zool. Garten 42, 1-16. Dusosrt, G. (1970): L’organisation spatiale et sociale de Muntiacus reevesi Ogılby, 1839 en semi- liberte. Mammalia 34, 331-355. Dusost, G. (1971): Observations &thologiques sur le Muntjak (Muntiacus muntjak Zimmermann, 1780 et M. reevesi Ogilby, 1839) en captıvite et semi-liberte. Z. Tierpsychol. 28, 387427. Dusost, G. (1980): L’ecologie et la vie sociale du Cephalophe bleu (Cephalophus monticola Thunberg), petit ruminant forestier afrıcain. Z. Tierpsychol. 54, 205-266. Dusost, G. (1983): Le comportement de Cephalophus monticola Thunberg et C. dorsalis Gray, et la place des Cephalophes au sein des ruminants. Mammalia 47, 141-177, 281-310. FEER, F. (1982): Quelques observations Ethologiques sur l’Hydropote de chine, Hydropotes inermis (Swinhoe, 1870) en captivite. Z. Säugetierkunde 47, 175-185. FRÄDRICH, H. (1964): Beobachtungen zur Kreuzung zwischen Schwarzrückenducker, Cephalophus dorsalis Gray, 1846, und Zebraducker, Cephalophus zebra Gray, 1838. Z. Säugetierkunde 29, 46-51. HALTENORTH, Th. (1963): Klassifikation der Säugetiere: Artiodactyla. In: KüKENTHAL, Hb. Zool. 8, 31, 1 (18). Berlin: de Gruyter. Horkıns, J. M. (1966): Observations on the rearıing and behaviour of young duikers in captivity. Nigerian Field 31, 118-131. Kıncoon, J. (1979, 1982): East Afrıcan Mammals. An Atlas of Evolution in Africa. Vol. 3. London and New York: Academic Press. LYDEkKER, R. (1924): The game anımals of India, Burma, Malaya, and Tibet. 2nd ed. London: Rowland Ward. PRATER, S. H. (1965): The book of Indian Mammals. 2e ed. Bombay Nat. Hist. Soc., and Prince of Wales Museum of Western India. Rauıs, K. (1970): Duikers from African forests to African plains. Anım. Kingd. 73, 18-23. Raııs, K. (1974): Scent Marking ın Captive Maxwell’s Duikers. In: The Behaviour of Ungulates and its relation to management, Ed. by Geist and Walther, rucn Publ., New Ser. 24, 114-123. Raııs, K. (1975): Agonistic behavior in Maxwell’s Duiker, Cephalophus maxwelli. Mammalıa 39, 241-249. Ras, K. (1984): Duikers. In: The Encyclopedia of Mammals. Ed. by Mac Donald. London and Sydney: George Allen and Unwin. 556-559. SCHWEERS, $. (1984): Zur Fortpflanzungsbiologie des Zebraduckers Cephalophus zebra (Gray, 1838) im Vergleich zu anderen Cephalophus-Arten. Z. Säugetierkunde 49, 21-36. WALTHER, F. (1968): Ducker, Böckchen und Waldböcke. In: Grzimeks Tierleben. Bd. 13, Zürich: Kindler. 342-367. WALTHER, F. (1979): Das Verhalten der Hornträger (Bovidae). In: KükENTHAL, Hb. Zool. 8, (54), 1-184. Berlin: de Gruyter. Adresse des auteurs: G. DuBost et F. FEER, Laboratoire d’Ecologie du Museum National d’Histoire Naturelle, 4, Avenue du Petit Chäteau, F-91800 Brunoy, France Z. Säugetierkunde 53 (1988) 48-54 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Systematics and chromosomes of the Indian gazelle, Gazella bennetti (Sykes, 1831) By CH. W. FurLey, H. Tıchy, and H.-P. UERPMANN Receipt of Ms. 15. 1. 1987 Abstract In 3 individuals of Gazella bennetti chromosome numbers of 2n = 50 (?) and 51 (@) were found. The structures ofthe X and Y1 chromosomes differ from other gazelle species. The Indian gazelle must not be classified as a subspecies of G. gazella or G. dorcas. Introduction The genus Gazella de Blainville, 1816, has always been a major challenge to the mamma- lian taxonomist. There is still room for subjectivity in the classification of some members of this group, not only at the subspecies but even at the species level. No common agreement has yet been reached on the number of genuine species within this genus, especially with regard to the “smaller gazelles” (which exclude the subgenus Nanger). For example, no final answer can be given to the question whether Gazella leptoceros should be regarded as a species of its own or whether ıt should be incorporated into Gazella subgutturosa (as proposed by Lange [1972]). Another gazelle with an uncertain taxonomic position is the Chinkara or Indian gazelle, Gazella bennetti (Sykes, 1831), which was included into Gazella gazella by some authors, whereas others considered it as belonging to Gazella dorcas (see below for further discussion). As will be shown in thıs article, neither of these views is correct. Some of the uncertainties about the classification of smaller gazelles are due to the fact that variation within the units commonly regarded as species, is almost as big as the total range of varıation within the remarkably uniform subgenus Gazella. There is such an overlap between intra- and interspecific varıation both in coloration and cranial morphol- ogy that geographical provenience of an individual specimen is often the most important character for its classification — which is certainly not the correct procedure for taxonomic identification. This ıs particularly true due to human interference with the distribution of gazelles in recent years. Gazelles are favorite pets all over the Middle East and North Africa and there ıs a flourishing, yet completely uncontrolled trade of live anımals over long distances. Thus, the locality where a specimen was obtained by a zoological collector may be far away from the place of origin of the anımal. In addition, the trade and keeping of gazelles in one’s backyard has an immanent danger of hybridization, which adds to the difficulties of classification. One possibility to decrease the difficulties of gazelle systematics ıs to establish the exact geographical ranges of defineable forms prior to the distortions caused by human interfer- ence. Bone remains of gazelles are common finds at most archaeological sites in the Middle East and North Africa. Once they can be identified beyond the genus level, these finds have a potential to determine the original range of the respective species. However, except for complete skulls or well preserved fragments thereof, specific identification of ısolated gazelle bones is still impossible. To establish criteria for the identification of posteranial bones (apart from the evaluation of differences in absolute size which may have shifted during the recent geological past), it is necessary to increase the number of well identified U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5301-0048 $ 02.50/0 Systematics and chromosomes of the Indian gazelle, Gazella bennetti 49 reference skeletons. Obviously, this necessity, namely correct taxonomic identification of the reference specimen, brings us back to the difficulties described above. This article is a first result of some work based on the attempt of one of us (H.P.U.) to build up a comparative collection of gazelle skeletons for palaeo-zoological research. Fairly large breeding groups of different gazelle species kept under the medical care of another of us (C.W.F.) at the zoo of Al-Ain (Abu Dhabi, U.A.E.) were a source not only of skeletal material, they also provided the living tissues to try a ‘biological’ determination of the taxonomic affinities of the respective populations. As a basic approach, some karyological research was started by the third member of our group (H.T.). It is mainly due to this last part of pur work that some contributions to the systematics of gazelles have resulted from this effort. Some gazelles brought from Pakistan to the zoo of Al Ain gave surprising results when tested for their chromosome numbers. Some conclusions on the systematic position of these gazelles will be based on a description of the animals themselves and on the evaluation of their karyotypes. Material and methods A group of gazelles captured during a hunting expedition in Pakistan was brought to the zoo of Al- Ain ın Abu Dhabi (United Arab Emirates) in 1984. Unfortunately, the exact geographical origin of the animals could not be investigated. It was obvious, however, that they were “chinkaras”, which is the local name for the Indian gazelle, Gazella bennetti. The general colour of the animals is a light fawn, almost isabelline. The stripe along the flank is brown, narrow at its origin on the shoulder but increasing in width as ıt extends along the lower part of the rump. The upper rim of this stripe is not well defined nor is there a well marked zone of lıghter coloration above the flank stripe. The lower border of the flank stripe ıs well marked against the yellowish colour of the ventrum which also extends along the inner thighs of the legs. Like the flank stripe, the pygal stripe is brown, fadıng into the fawn colour of the back, but contrasting well with the whitish patch on the inside of the thighs. The pygal stripe extends down the sides of the tail, ending ın the black of the distal portion of the tail. Other spots of dark coloration are just above the hoofs, particularly between the two digits, and the carpal brushes. The facial stripes are also dark brown in colour, extending from the anterior corner of the eyes finishing just distal to the upper lips. Above and below these stripes, as well as around the eyes and the mouth, the colour ıs very light. The nasal area between the facıal stripes ıs fawn with a typical dark nose blotch of varying extent. In some anımals it is an ıll defined brown spot only. Usually ıt ıs almost black in the middle with shades of brown towards the edges. Between the eyes, the colour is reddish-fawn. Darker eyebrow stripes reach from the eyebrows to the lateral edges of the horns, fading toward the base of the large ears. The frontal region is lighter again, though not much different from the general colour of the animal. Compared to most other gazelle species, the markings of our anımals are fairly unconspicuous. This is a known feature of Gazella bennetti. In comparison to typical Indian anımals of this species, our animals are less reddish, more greyish ın total appearance. Specimens from Baluchistan, originally considered to be a separate species (Gazella fuscıfrons, Blanford, 1873), have a dark coloration of the forehead, not present in our anımals. The description available for Gazella bennetti christyı Blyth, 1841, which is the subspecies inhabiting the intermediate zone between the two extremes, is t00 vague to be applied. The “silvery drab-brown” (Grovzs 1985) of this subspecies might in fact apply to our anımals. However, they could also be called “rich tobacco-brown” which — according to GROVES (1985) - is the coloration of the unnamed subspecies inhabiting the Salt Range and Punjab areas of northeastern Pakistan and northern India. Thus, the better known subspecies bennetti and fuscifrons can both be excluded, whereas the two less well known races of Pakistan are both possible identifications for the anımals used here. For karyotyping, the blood of three anımals was tested: An adult female and its male calf, and an unrelated adult male. Blood samples of 3-5 ml were sent in a heparinized form from Al Ain to Tübingen by air courier. The time span between taking the blood samples and the beginning of cell cultivation was 48 to 60 hours. Cell culture was carried out by isolating Iymphocytes over a ficoll paque gradient, and stimulation of mitoses with phytohemagglutinine in RMPI 1640 medium with 15 % fetal calf serum. After 72 hours of culture at 37°C, 5 % CO,, mitoses were arrested with colchicine. Metaphase chromosome spreads were prepared after hypotonic treatment and fixation in methanol/acetic acıd (3:1) by routine air dry techniques. For identification, the metaphases were stained with orcein, and for the characterisation of constitutive heterochromatine, the C-banding technique was used (SUMNER 1972). 50 Ch. W. Furley, H. Tichy and H.-P. Uerpmann Fig. 1. Young male of Gazella bennetti from Pakistan in the quarantine station of Al Ain Zoo Results For the three anımals under consideration we found a diploid chromosome number of 2n = 50 (?) and 2n = 51 (S). The autosomes are formed by 4 pairs of individually distingui- shable metacentric and 20 pairs of telo- to acrocentric chromosomes. The last ones have very minute p-arms, and only in metaphases with long, slightly condensed chromosomes could the p-arms be identified. Satellites were sometimes visible in these metaphases at the end of the q-arms of some of the largest acrocentric chromosomes. No further grouping of the autosomes was possible after orcein staining (fig. 2). The heterosomes are 2 submetacentric X-chromosomes in the female, and one X, a submetacentric Y1 and an acrocentric Y2 chromosome in the male. The X is the largest of allchromosomes, containing about 14 % of the total chromosome length. Its p:q ratio is just 1:2. The Y1, also with a p:q ratio of 1:2, was individually identifiable even after orcein staining alone. Y2 could not be distinguished from other acrocentrics of sımilar length after orcein staining. | C-banding reveals constitutive heterochromatine at the kinetochore regions of all the autosomes. Two pairs of telocentric chromosomes exhibited heterochromatine at their telomeres. Most striking was the observation of constitutive heterochromatine accumula- ted along the total length of the p-arms of the X and Y1 chromosome (fig. 3). a ER E; Systematics and chromosomes of the Indian gazelle, Gazella bennetti u Sl Fıg. 2. Karyotype of Gazella bennetti, orcein stained 52 Ch. W. Furley, H. Tichy and H.-P. Uerpmann Fig. 3. C-banded chromosomes of a female and a male individual of Gazella bennetti. The arrows point to the heterochromatic arms of the X chromosomes and to the Y1 chromosome (short arrow) Discussion There are three points of interest in comparing the karyotype ot Gazella bennetti with published karyotypes of other members of the genus Gazella (Hsu and BENIRSCHkE 1967/ 77; WURSTER 1972; EFFRON et al. 1975; BENIRSCHRE et al. 1984): 1. the chromosome number, 2. the shape of the X and Y1 chromosomes with a p:q ratio of 1:2, and 3. the amount of constitutive heterochromatine in the p-arms of X and Y1 which all differ greatly from those previously described. The high chromosome number of 2n = 50 2/51 & of Gazella bennetti ıs only outdone by 2n =58 ın both sexes of Gazella thomsoni (Hsu and BENIRSCHEE 1968). The autosomes of this species are all telo- to acrocentric. Of the other gazelles. Gazella dama with 3840 chromosomes in the female and 39-40 in the male (BENIRSCHkE et al. 1984) ıs nearest to Gazella bennetti. Varying chromosome numbers due to Robertsonian fusions/fissions in a zoo popula- tion of Gazella soemmeringi were published by BENIRSCHRE et al. (1984). In the case of Gazella bennetti, multiple fissions of different metacentrics may have lead to the numbers found in our investigations. Such a process could even have started from a chromosome number as low as the ones found in Gazella grantı or Gazella subgutturosa. The 14 pairs of metacentrics of the latter species could be translated — by 10 fissions — into 4 remaining metacentrics and 20 acro- to telocentric pairs as observed in the investigated anımals. A detailed analysis of the chromosomes by other banding techniques, which gives further evidence for this hypothesis, will be published separately. Because of the small number of individuals available, no comment is possible on the natural varıation in chromosome numbers in Gazella bennett:. Ä The two other points of interest characterising the karyotypes of the investigated anımals are of importance in considerations on the systematic position of Gazella bennett:. An X chromosome of comparable length with a similar p:q ratio is present in Gazella subgutturosa, Gazella leptoceros, Gazella gazella, Gazella granti, and Gazella dama. Only for the first three species have there been reports that the Xp behave heterochromatic. A Systematics and chromosomes of the Indian gazelle, Gazella bennetti 53 meta-submetacentric Y chromosome is found in Gazella soemmeringi (BENIRSCHRE et al. 1984), but nothing was published on its content of heterochromatin. Constitutive hetero- chromatine in a Y chromosome (either Y1 or Y2) has only been demonstrated for the acrocentric Y1 of Gazella subgutturosa (Hsu and BENIRSCHKE 1977; BENIRSCHEE et al. 1984). ELLERMAN and MoRrRrIson-ScotT (1951), followed by HALTENORTH (1963) and ROBERTS (1977) grouped Gazella bennetti as a subspecies wıth Gazella gazella. GROVES (1969), tollowed by Lance (1972) and CorBEr (1978) placed it with Gazella dorcas. Only recently Grovzs (1985) has revised his views due to the accumulating evidence for an independent position of the Indian gazelle. From the cytological criteria demonstrated above, the investigated anımals are too different from both gazella and dorcas to be interpreted as being just a varıation of the karyotype of one of these species. As has been shown, the closest accordance exists with Gazella subgutturosa. This is surprising only if this species is really regarded as belonging to a different subgenus (Trachelocele Ellerman and Morrison-Scott, 1951). The existence of a throat-swelling in males of the goitred gazelle during the breeding season — which is the character used by ELLERMAN and MORRISON-ScoTT (1951) to define this subgenus — does not appear to be a good character for a subgeneric separation of Gazella subgutturosa. In any case, the remaining cytological differences such as the number of necessary Robertsonian fissions or fusions and the previously unreported form of the Y1 chromosome make it unlikely that our Gazella bennetti is too closely related even to the latter species. The karyological differences point to the necessity of establishing Gazella bennetti as a species in its own rights. Since Gazella bennetti and Gazella subgutturosa have an overlap in distribution and since there are no intergrading populations, there would be no doubt that they are separate species even without evidence of karyological differences. Gazella bennetti was included into Gazella dorcas because of similarities in cranial morphology. Having to exclude it now, also changes the range of variation of this last species. Without bennetti, the remainder of the Dorcas gazelles becomes more uniform, which in reverse affects the status of other marginal groups. Particularly the other eastern forms with fairly straight horns, like the Saudi gazelle, Gazella dorcas sandiya Carruthers and Schwarz, 1935, or Pelzeln’s gazelle, Gazella dorcas pelzelni Kohle, 1886, are candida- tes for exclusion from Gazella dorcas. Together with the gazelles from the Red Sea islands (GrovEs 1983) they might rather form a complex wıth Gazella bennetti. It would be interesting to compare the karyotypes of these taxa. It is obvious from our results that chromosome studies will add further pieces of evidence to the yet uncomplete understanding of evolution and systematics of the gazelles. Future schemes should at any rate separate the forms according to the occurrence of a second Y chromosome ın the males, which seems to be a particular evolutionary feature of most gazelles (WaHrman et al. 1973). Thomson’s gazelle, and if they are conspecific (Grovss 1985) the Red-fronted and Heuglin’s gazelle as well, would have to be excluded from the gazelle genus on thıs basıs. Among the gazelles with a double Y chromosome, the subgenus Nanger will probably remain a useful subdivision, whereas Trachelocele in the sense of ELLERMAN and MORRISON-SCOTT (1951) will not. It is obvious on morphological grounds that Gazella subgutturosa ıs related to Gazella leptoceros (LANGE 1972). Our studies have revealed karyological affinities between subgutturosa and bennetti. Future studies will be necessary to investigate their morphological relations. Finally, one important remark has to be added: regional aspects must be considered carefully in future chromosome studies of gazelles. As in the case of Gazella bennetti, local populations — thought to belong to a more widespread species - may turn out to be independent taxa. It may later be possible to relate karyotypes and geographical distribu- tion. Thus, the geographical origin of the anımals under study, even if as unprecise as in our case, must be published together with the karyological results. This is, for example, 54 Ch. W. Furley, H. Tichy and H.-P. Uerpmann not the case in the „Chromosome Atlas” (Hsu and BENIRSCHKE 1967/77), from where basic information had to be used in this study as well. Acknowledgements The efforts to build up a comparative collection of gazelle skeletons were supported by the “Sonderforschungsbereich 19 — Tübinger Atlas des Vorderen Orients” of the DFG and by the University of Tübingen. We wish to thank these institutions for their support. Zusammenfassung Systematik und Chromosomen der Indischen Gazelle, Gazella bennetti (Sykes, 1831) Bei 3 Individuen von Gazella bennetti wurden Chromosomenzahlen von 2n = 50 (?) und 51 (8) gezählt. Aufbau und Gestalt der X- und Y1-Chromosomen unterscheiden sich von denen anderer Gazellenarten. Die Indische Gazelle darf nicht als Unterart von G. gazella oder G. dorcas klassifiziert werden. Literature BENIRSCHKE, K.; KUMAMOTO, A. T.; OLseEn, J. H.; WıLLıams, M. M.; OoSTERHUISs, J. (1984): On the chromosomes of Gazella soemmeringi Cretzschmar, 1826. Z. Säugetierkunde 49, 368-373. CoRBET, G. B. (1978): The Mammals of the Palaearctic Region. A Taxonomic Review. London: Brit. Museum (Nat. Hist.) and Ithaca: Cornell Univ. Press. EFFRON, M.; BoGArRT, M. H.; KumamoTo0, A. T.; BENIRSCHKE, K. (1976): Chromosome studies in the mammalian subfamily Antilopinae. Genetica 46, 419444. ELLERMAN, J. R.; MORRISON-ScOTT, T. C. S. (1951): Checklist of Palaearctic and Indian mammals 1758 to 1946. London: Brit. Museum (Nat. Hist.). Grovss, C. P. (1969): On the smaller gazelles of the genus Gazella de Blainville, 1816. Z. Säugetierkunde 34, 38-60. Grovezs, C. P. (1983): Notes on the Gazelles IV. The Arabian Gazelles collected by HEMPRICH and EHRENBERG. Z. Säugetierkunde 48, 371-381. Grovss, C. P. (1985): An Introduction to the Gazelles. Chinkara 1, 4-16. HALTENORTH, TH. (1963): Klassifikation der Säugetiere: Artiodactyla. Handbuch der Zoologie 8/32, 1-167. Hsu, T. C.; BENIRSCHKE, K. (eds.) (1967/77): An Atlas of Mammalian Chromosomes. New York: Springer. vol. 2, fol. 93 (1968) - G. thomsoni; vol. 8, fol. 394 (1974) — G. dorcas; vol. 9, fol. 439 (1975) — G. grantı; vol. 10, fol. 503 (1977) - G. leptoceros; vol. 10, fol. 504 (1977) - subgutturosa. LANGE, J. (1972): Studien an Gazellenschädeln. Ein Beitrag zur Systematik der kleineren Gazellen, Gazella (De Blainville, 1816). Säugetierkl. Mitt. 20, 193-249. ROBERTS, T. J. (1977): The Mammals of Pakistan. London: Ernest Benn. SUMNER, A. T. (1972): A simple technique for demonstrating centromeric heterochromatın. Experi- mental Cell. Res. 75, 304-306. WAHRMAN, J.; RICHLER, C.; GOITEIN, R.; HOROWITZ, A.; MENDELSOHN, M. (1973): Multiple sex chromosome evolution, hybridization, and differential X-chromosome inactivation in gazelles. Chromosomes Today 42, 434435. WURSTER, D. H. (1972): X-chromosome translocations and karyotypes in bovid tribes. Cytogenetics 1154197207 Authors’ addresses: Chris W. FURLEY, Rufflans, Bekesbourne Lane, Bekesbourne, Nr. Canterbury, Kent CT3 1XB, England; Dr. HERBERT TıcHhy, Max-Planck-Institut für Biologie, Spemannstr. 34, D-7400 Tübingen, FRG, and Dr. Dr. habil. Hans- PETER UERPMAnN, Institut für Urgeschichte, Schloss, D-7400 Tübingen, FRG Z. Säugetierkunde 53 (1988) 55-56 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-346 NAISSEINISE FIABIEEIETTESKURZMITELEILUNGEN A note on aquatic and aerial vision in Odontocetes By A. D. G. DrarL Netherlands Institute for Sea Research, Den Burg, Texel, The Netherlands Receipt of Ms. 17. 2. 1987 The paper by NeuHaus, which recently (1986) appeared in this journal, contains an over simplified analysıs which casts doubt on some of its conclusions. The author presents calculations, based on the anatomy of the eye of the Beluga (White Whale, Dephinapterus leucas), from which it appears that the eye is approximately in focus (1.e. well focused on objects at infinity), when it is used under water. Lacking a mechanısm for accommodation, it follows that in air, where refraction by the cornea becomes effective, the Beluga must be very near-sighted. This conclusion, however, conflicts with the Beluga’s apparent use of vision in air. The author proposes a resolution to this conflict by first assuming that the image formed by the eye under water lies on a flat surface slightly forward of the retina and perpendicular to the optical axis. This paraxial image plane intersects the retina slightly to the sıde of the optical axis and thus avoids the “blind spot” resulting from the axial location of the optic nerve. Carrying this ıdea further, the author then reasons that the Beluga could be in focus in air also. That is, as the paraxial image plane moves much closer to the lens, with the addition of the refractive power of the cornea in air, it will again intersect the retina at some greater annular distance from the optical axis. The reader is led to believe that irrespective of the viewing medium or distance the non-focusing eye of the Beluga will be in focus somewhere on the retina (with the underlying assumption that these regions of the retina are capable of good resolution). These ideas were not supported by new data and are inconsistent with a more general geometrical optics analysıs. Fundamentally, the assumption that the image is focused on a flat surface is in error and the inferences based on this assumption are also incorrect. Images formed by simple optical systems are projected on a curved surface (which might well be the very reason for the curved shape of the vertebrate retina). This phenomenon is known as the Petzval curvature of the image. For a plane object at infinity and a system of k refracting surfaces, the radıus of curvature of the final image in the absence of oblique astıigmatism and spherical aberration can be calculated with the following equation (LONGHURST 1967, p. 359-360): RER (ei,1n) Rx en N;+1 n; in which R; is u radius of curvature of the image in the k"? medium, r; is ii: radıus of curvature of the i® refracting surface and n; is the refraction index after the i'® refracting surface. Applying this equation to the eye of the Beluga as described by NEUHAUS, for aquatic vision a Petzval curvature of the retinal image is 10.0 mm. In view of the uncertainty of the values used in the calculation, the result may be considered to be in reasonable agreement with the curvature of the retina, which has, according to NEUHAUS, a radıus of 11.3 mm. We may conclude that the Beluga eye under water is in focus over the entire retinal surface (as could be expected a priori). This near coincidence of the retina and under water image eliminates the need to seek a mechanısm to avoıd the axially located U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5301-0055 $ 02.50/0 56 A. D. G. Dral “blind spot” caused by the optic nerve. The other key feature of the NEuHAus Beluga eye model is that it would be focused in air. However, the aerıal Petzval image curvature of the NEUHAUS eye model indicates that the anımal is hopelessiy near- sighted i in air. One would expect toothed whales to have good far vision in air because echo location does not provide remote sensory information in this medium and good near vision in water because the oceans attenuate light rapıdly with distance. In this regard the eye model presented by Rıvamonte (1976) offers an alternative explanation which agrees with the behaviorally measured near-sightedness in water and far-sıghtedness in air results of HERMman et al. (1975). In this model of the eye of the bottle-nosed dolphin (Tursiops truncatus), the lens is assumed to be bi-focal, the core having an appreciably higher refractive index than the periphery. The lens core would be primarily functional in aquatic vision, allowing focused near vision under water. By virtue of the peculiar shape of the pupil and location of the lens, in air the image forming light would mainly pass through the less refractive peripheral part of the lens. The combination of a less refractive lens periphery and highly refractive cornea allows focused far vision in air. As argued elsewhere (Drau 1985), the elements of this model fit nıcely into the known facts of ocular anatomy, enhancing its credibility. Acknowledgements I am indebted to Dr L. Maas (NIOZ, Texel) for enlightening me on the computational aspects of this paper, and to Mr. A. RıvamonTE (Army Primary Standards Laboratory, USA) for his correction of the manuscript. References Drar, A. D. G. (1985): Amphibious vision in dolphins. In: Vertebrate Morphology. Ed. by H.-R., Duncker and FLEISCHER, G. Fortschritte der Zoologie 30, 707-709. Herman, L. M.; PEAcock, M. F.; YUNkER, M. P.; Mapsen, C. J. (1975): Bottlenosed dolphin: double slit pupil yields equivalent aerıal and underwater diurnal acuity. Science 189, 650-652. LoNGHUußsT, R. S. (1967): Geometrical and physical optics, 2nd ed., Longman: London. NEUHAUS, W. (1986): Die Bedingungen für das Sehen des Weißwals, Delphinapterus leucas Pall., in Wasser und Luft. Z. Säugetierkunde 51, 266-273. RIVAMONTE, L. A. (1976): Eye model to account for comparable aerial and underwater acuities of the bottlenose dolphin. Neth. J. Sea Res. 10, 491-498. Author’s address: A. D. G. Drar, Netherlands Institute for Sea Research, P. ©. Box 59, NL-1790 AB Den Burg, Texel, The Netherlands Z. Säugetierkunde 53 (1988) 57-58 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Zur Frage der Sehfähigkeit von Delphinapterus leucas in Wasser und in Luft Von W. NEUHAUS Eingang des Ms. 25. 8. 1987 In seiner Mitteilung über das Sehen der Zahnwale in Wasser und Luft weist A. Drau (1987) darauf hin, daß meine Berechnung über das Sehen des Weißwals ın Luft, wenn die nach PErzvaL berechnete Bildkrümmung berücksichtigt wird, fehlerhaft wird und zu extremer Kurzsichtigkeit führt. Er selbst vertritt die Auffassung von RıvamoNTE (1976), die eine ganz andere Grundlage als meine Untersuchung hat. Aus der klaren Zeichnung von PıLLErı (1964) konnten die morphologischen Größen des Weißwalauges entnommen werden, die allerdings nicht vollständig mit einigen vom gleichen Autor mitgeteilten Zahlenwerten übereinstimmen (NEUHAUS 1986). Die Brechungsexponenten wurden in Anlehnung an die Verhältnisse des menschlichen Auges erschlossen, der Wert für die Linse jedoch aus den von MATTHIESSEN (1886) gemessenen Brechungsexponenten für die Linse verschiedener Zahnwale und anderer Cetaceen gemit- telt. Diese Größe ist 1,60, demnach bis zu einem gewissen Grade willkürlich. Ändert man sie geringfügig auf 1,59, so wird die Brennweite der Linse ein wenig verlängert und die Werte für die Bildkrümmung nach der PETZvar-Formel verändert. Man erhält dann wie in meiner früheren Untersuchung zwei kreisförmige Zonen scharfen Sehens auf der Retina, eine für das Sehen in Wasser, die andere für das Sehen in Luft (Abb. 1). Pırrerr stellte im histologischen Teil seiner Arbeit im Unterschied zu Drau keine Unterschiede ın der Verteilung der Sinneszellen der Retina fest. Die vorstehende Ausführung entspricht bei Berücksichtigung der Korrektur dem Ergebnis meiner früheren Untersuchung. A. Drau (1987) vertritt die von C. A. RıvamonTeE (1976) entwickelte Auffassung. Im Wasser ist danach das Auge der Zahnwale bei geöffneter Pupille sehtüchtig. Bewegen sie sich in Luft, so werden die Pupillen wegen der stark gesteigerten Helligkeit bis auf schlitzförmige Öffnungen an den Rändern geschlossen. Das hier einfallende Licht bildet nach der These die Umgebung auf der Retina scharf ab, weil der Brechungsindex und damit die Brechkraft der Linse in den Randbezirken niedriger ist als in der Mitte. Hierdurch soll für das Gesamtauge der Unterschied in den Brechungsindices zwischen Wasser und Luft kompensiert werden. Der Schichtenaufbau der Linse bringt es bei den Säugetieren mit sich, daß die Bre- chungsindices allmählich von außen nach innen, und zwar am meisten in den äußeren äquatorialen Schichten (GROTHUISEN 1929) zunehmen. Bei großer Helligkeit treffen einfallende Lichtbündel alleın auf die peripheren Linsenab- schnitte mit der stärksten Gradation der Brechkraft. Die Strahlen eines Gegenstandspunk- tes, z.B. in 2m Entfernung, divergieren im Auge bis zur Weite der Randschlitze der Pupillen (etwa 0,6-0,8 mm). Die der Linsenmitte näheren Strahlen werden stärker gebrochen als die äußeren, denn dıe erwähnte Strecke kann bei einem Linsendurchmesser um 6 mm nicht vernachlässigt werden (Abb. 2). Unter diesen Bedingungen kann es nicht zu einer scharfen Abbildung auf der Retina kommen. Auch der verhältnismäßig langsame Pupillenreflex könnte bei den schnellen und oft zentimetergenau gezielten Sprüngen der Delphine für die visuelle Orientierung in Luft (nach Rıvamonrte) hinderlich sein. Die Ausführungen zeigen, daß die von mir vertretene Auffassung über das Sehvermögen des Weißwals in Wasser und Luft aufrechterhalten werden kann. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5301-0057 $ 02.50/0 58 W. Neuhaus J & E Ll za ee a , Br Tan Z\e Sr X Rw 7 A yG Y% a % Ya a Ga RE , Lw“ Abb. 1. Schema der Optik des Auges von Delphinapterus leucas. C = Cornea, I = Iris, Kw = Pro- jektion der Kreiszone für scharfes Sehen im Wasser, Kl = dasselbe für das Sehen in Luft, L = Linse, Lw = Einstrahlendes Lichtbündel im Wasser, Ll = dasselbe für das Sehen in Luft, Rw = Petzval-Rand ın Wasser, Rl = dasselbe ın Luft Abb. 2. Schema des Strahlenganges im Auge von Delphinapterus lencas, wenn alleın Randschlitze der Pupille offen sind. I = Iris (Randschlitz), L = Linse, R = Retina, — einstrahlendes Lichtbündel, --- hypothetischer Strahl nach der These von RIVAMONTE Literatur Drar, A. D. G. (1985): Amphibious vision in dolphins. Fortschr. d. Zoologie 30, 707-709. — (1987): A note on aquatic and aerıal vision in Odontocetes. Z. Säugetierkunde 52, 55-56. GROTHUISEN, G. (1929): Abbildungstiefe des Auges. Hdb. morm. u. pathol. Physiologie XII, 1. MATTHIESSEN (1886): Über den physikalisch-optischen Bau des Auges der Cetaceen und der Fische. Arch. ges. Physiol. NEUHAUS, W. (1986): Die Bedingungen für das Sehen des Weißwals, Delphinapterus lencas Pall., in Wasser und Luft. Z. Säugetierkunde 51, 266-273. | PiLLert, G. (1964): Zur Morphologie des Auges vom Weißwal Delphinapterus leucas. Hvalrädets Skrifter. 47, 1-16. RIVAMONTE, L. A. (1976): Eye model to account for comparable aerial and underwater acuities of the bottlenose dolphin. Neth. J. Sea Res. 10, 491-498. Anschrift des Verfassers: Prof. Dr. WALTER NEUHAUS, Zoologisches Institut und Zoologisches Museum, Universität Hamburg, Martin-Luther-King-Platz 3, D-2000 Hamburg 13 Z. Säugetierkunde 53 (1988) 59-60 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 On the pupping period of Grey Seals, Halichoerus grypus (Fabricius, 1791), reproducing on a shoal near the Island of Terschelling, the Netherlands By L. ’r HarT, A. MOoESKER, L. VEDDER and P. J. H. van BREE Harbour Seals Rehabilitation and Research Centre, Pieterburen, The Netherlands Receipt of Ms. 3. 4. 1987 Although the species must have been common in prehistoric times in view of the many remnants found during excavations, Grey Seals were unknown in the Netherlands until the winter of 1955-56, when a specimen was observed in the vicinity of the Island of Texel (KrısTENSEN 1957). From that winter on many Grey Seals were found on the coast of the Netherlands (HıckLıng 1962; Van HAAFTEN 1974), almost certainly all originating from the large colony in the Farne Islands in the North Sea, opposite the border between England and Scotland. Most of the specimens were very young anımals and often in a poor condition. When found these young seals were mostly taken to the Harbour Seals Rehabilitation and Research Centre at Pieterburen, while some were taken to the Nature Education Centre on Texel. After recovery, they were set free again in the North Sea or the adjacent Wadden Sea. Thus, over the years non-reproducing groups of Grey Seals were formed lıving in the coastal water of the Netherlands. At present one of these groups ıs living on and near “Engelse Hoek’, a high Iying shoal (53° 20’ N, 5° 10’ E) off the Friesian Island of Terschel- ling. In the second week of January 1985 pilots of military aircraft reported that they had spotted two female Grey Seals with suckling young. Once this became known, special attention was given to the fact that seals might be born on the shoal. In the last week of January 1986 pilots again reported having seen two females with young. In the fırst week of February 1987 three newborn Grey Seals were seen lyıng on the shoal, while 10 days later two more were spotted. This shoal ıs a part of the Wadden Sea that remains dry most of the time, being flooded only in case of high northwesterly winds (force 7 de Beaufort and stronger). That is probably the reason why a number of Grey Seals chose this shoal for their habıtat. From 1980 to 1986 a total of 55 stranded Grey Seals were taken into the Pieterburen centre, recovered and set free again on this shoal. From observations we know that at present the shoal accommodates a colony of some 45 Grey Seals. The two Grey Seals born in January 1985 disappeared after heavy weather. One was probably found on the Friesian coast and could be rehabilitated in the Pieterburen centre. The two Grey Seal pups born in the last week of January 1986 also suffered bad luck. They disappeared in stormy weather, one being later found dead on the shore of Terschelling. The Grey Seal pups born in February 1987 numbered five. Three were found on the shore of Terschelling and survived thanks to treatment in the Pieterburen centre. The other two grew up on the shoal. That Grey Seals should settle and breed in the coastal waters of the Netherlands was to be expected, but that the pupping period would be in the second and third week of February came as a surprise. After all, the breeding season of the colony of Grey Seals in the Farne Islands, where most of the Dutch specimen come from, starts around the middle of October and lasts until around the middle of December, with a peak by 7 November U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5301-0059 $ 02.50/0 60 L. t’Hart, A. Moesker, L. Vedder and P. J. H. van Bree SERIT OCT. NOV. DEC. JAN. FEB. | MARCH. | APRIL. S.WALES ORKNEYS FARNES Fig. 1. Pupping seasons and peak times of some Grey Seal colonies. After Kıng (1983) (HEwER 1974; Kıng 1983). That in the first year Grey Seals had pups on the shoal near the Island of Terschelling in January could be explained by the fact that this involved young females. Young Grey Seals have been known to give bırth in the spring (BACKHOUSE and HEewer 1957). That the young near Terschelling should be born in a later period in 1986 and 1987, however, was completely unexpected. The pupping season of the Grey Seals in the Netherlands looks more like that of Halıchoerus grypus ın the Baltic (Fig. 1). In this short article we will not try and furnish an explanation of the phenomenon, but we only want to draw attention to it ın order that it may be studied in the coming years. To conclude this note, we would like to express our gratefulness for the information we received from pilots of Leeuwarden AFB, and from Mr. HEssEL WIEGMAN. Literature BACKHOUSE, K. M.; HEwER, H.R. (1957): A note on spring pupping in the Grey Seal. Proc. zool. Soc. London 128, 593-596. HAAFTEN, ]J. L. van (1974): Zeehonden langs de Nederlandse kust. Wetenschappelijke Medd. KNNV 101, 1-36. HEwER, H.R. (1974): British Seals. London: Collins. Hickuing, G. (1962): Grey Seals and the Farne Islands. London: Routledge and Kegan Paul. Kıng, J. E. (1983): Seals of the world. Oxford: Oxford University Press. KRISTENSEN, 1. (1957): Een Grijze Zeehond (Halichoerus grypus) bij Texel. Lev. Natuur 60, 62-63. Authors’ address: Drs ALBERT MOESKER, Zeehondencreche, Hoofdstraat 94a, NL-9968 AG Pieter- buren, The Netherlands BEKANNTMACHUNGEN Einladung Die 62. Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde e.V. findet von Sonntag, dem 2. Oktober, bis Donnerstag, dem 6. Oktober 1988, in Münster statt. Vorläufiges Programm Sonntag, 2. Oktober: Anreise 19.00 Uhr: Zwangloser Begrüßungsabend im Zoo-Restau- rant (Sentruper Straße) Montag, 3. Oktober: 9.00 Uhr: Begrüßung und Eröffnung der Tagung durch den 1. Vorsitzenden im Vortragssaal des Westf. Mu- seums für Naturkunde, Sentruper Straße 285 9.30 Uhr: Hauptvortrag zu dem Schwerpunkt „Ethologie“, anschließend Kurzvorträge und Poster-Demon- stration 17.00 Uhr: Mitgliederversammlung 19.00 Uhr: Gemeinsames Abendessen im Museum Dienstag, 4. Oktober: 9.00 Uhr: Hauptvortrag zu dem Schwerpunkt „Energie- haushalt“, anschließend Kurzvorträge 14.00 Uhr: Hauptvortrag und Kurzreferate zu dem Schwer- punkt „Verbreitung“ 19.00 Uhr: Filmabend Mittwoch, 5. Oktober: 9.00 Uhr: Diskussionsrunde „Der Artenschutz und die Ge- setze“ 14.00 Uhr: Kurzvorträge, Führung durch den Zoo (Dr. G. RUEMPLER und Dr. K. KaIser) Donnerstag, 6. Oktober: Exkursion durch das westliche Münsterland (Feuchtwiesen, „Wildpferde“, Wasserschlösser) Alle Interessenten sind zu der Tagung herzlich eingeladen. Neben den angeführten Schwerpunkten werden wir auch diesmal wieder der Vielfalt der säugetierkundlichen Arbeitsgebiete Rechnung tragen (Kurzreferate und Poster-Demonstrationen). Das Programm mit der Vortragstolge wird allen Mitgliedern und auf Anfrage auch Nicht-Mitgliedern rechtzeitig vor der Tagung zugesandt. Falls besondere persönliche Einladungen gewünscht werden, wird gebeten, sich an den 1. Vorsitzenden Prof. Dr. E. KULZER, Institut für Biologie III, Auf der Morgenstelle 28, D-7400 Tübingen, zu wenden. Bitte melden Sıe dıe Vorträge, die nicht länger als 15 Minuten dauern sollen, und Poster-Demonstrationen für diese Tagung möglichst frühzeitig, spätestens aber bis zum 30. Aprıl 1988 beim Geschäftsführer der Gesellschaft, Prof. Dr. U. Sc#miDT, Zoologi- sches Institut, Poppelsdorfer Schloß, D-5300 Bonn 1, an. Fragen zum Tagungsort und zur Organisation sind zu richten an: Dr. M. BERGER, Westf. Museum für Naturkunde, Sentruper Str. 285, D-4400 Münster, Tel. 02 51/8 20 84. 62 Bekanntmachungen Ausschreibung des Förderpreises der Deutschen Gesellschaft für Säugetierkunde 1988 Die Deutsche Gesellschaft für Säugetierkunde schreibt den Förderpreis in Höhe von 3000,- DM als Anerkennung für hervorragende wissenschaftliche Leistungen junger Forscher aus. Voraussetzung ist eine im Druck vorliegende Arbeit aus den Gebieten Phylogenie und Systematik, Verbreitung, Ethologie, Ökologie und Populationsbiologie der Säugetiere. Die Arbeit muß in den drei vorausgehenden Kalenderjahren erschienen sein. Autoren dürfen nicht älter als 33 Jahre sein. Bewerbungen oder Vorschläge für den Förderpreis sind zu richten an die Geschäfts- stelle der Deutschen Gesellschaft für Säugetierkunde, Prof. Dr. Uwe ScHmiDT, Zoologi- sches Institut, Poppelsdorter Schloß, D-5300 Bonn 1, unter Beifügung von 5 Sonder- drucken. Der Jury gehören Wissenschaftler verschiedener Hochschulen und Universitäten und Mitglieder der Gesellschaft an. Der Preis wird bei der Eröffnung der Jahresversammlung der Gesellschaft in Münster (2. 10.-6. 10. 1988) überreicht. Internationale Kommission für Zoologische Nomenklatur The following Opinion has been published by the International Commission on Zoologi- cal Nomenkclature in the Bulletin of Zoological Nomenclature, vol. 43, part 4 on 11 December 1986: Opinion No. 1419 (p. 328) Democricetodon Fahlbusch, 1964 (Mammalıa, Rodentia): Democricetodon crassus Freudenthal, 1969 designated as type species. BUCHBESPEREECHUNKGEN Kress, A.; Mıtrıan, J.: The Female Genital Tract of the Shrew Crocidura russula. Advances in Anatomy, Embryology and Cell Biology. Vol. 101. Berlin-Heidelberg- London-Paris-Tokyo: Springer 1987. 76 pp., 31 figs. DM 63,-. ISBN 3-540-16942-3 Die Abhandlung bietet eine umfassende und sorgfältige Dokumentation des weiblichen Genitaltraktes der Haus-Spitzmaus, Crocidura russula im elektronenoptischen Bereich. Der Untersuchung liegen Befunde an 38 Individuen zugrunde, die durch Einzelbeispiele (Neomys, Sorex, Suncus) ergänzt werden. Die feinstrukturellen Befunde an Bursa ovarıca, Tuben, Uterus, Epoophoron, Cervix uteri und Vagina werden unter Berücksichtigung funktionellen und strukturellen Wandels (saisonal, Periode des Sexualcyclus) dargestellt. Damit liegt eine geschlossene Bearbeitung der Feinstrukturen des weiblichen Genitaltraktes für einen Insectivoren vor. Ausgezeichnete EM-Photos erläutern den ext Vergleichende Hinweise auf Monotremata, Marsupialia und Chiroptera beschränken sich auf Auswertung des Schrifttums. Eine Reihe von Mißsverständnissen, wie etwa die Behauptung „The Marsupials are thought to be the oldest placental mammals“, bedürfen der Korrektur, wenn auch die Autorinnen zu dem Schluß kommen, dafs die Fortpflanzungsmechanismen der Beuteltiere keineswegs einfacher sind als die der Eutheria und einen alternativen Modus darstellen. Dies ist nicht neu. Unglücklich ist auch die Formulierung, mit der die Soricidae den „more recent species“ gegenüberge- stellt werden. Das Vorkommen von Plesiomorphien ist kein Beweis für direkte Abstammungsver- wandtschaft. Der Wert der eigenen Befunde wird durch diese Bemerkungen allerdings nicht beein- trächtigt. D. STARcK, Frankfurt/M. Buchbesprechungen 63 SCHOPPE, R.: Die Schlafmäuse (Gliridae) in Niedersachsen. Lebensraum und Verbrei- tung von Siebenschläfer, Gartenschläfer und Haselmaus. Naturschutz und Landschafts- pflege in Niedersachsen, Beiheft, Heft 14. Hannover 1986. 52 S. DM 7,-. ISBN 3-922321-36-4 Verbreitung und Lebensraum von Siebenschläfer, Gartenschläfer und Haselmaus in Niedersachsen werden aufgrund von 575 Funden (247 neu, 328 aus der Literatur) beschrieben. 65 % der Angaben betreffen Glis, 13 % Ehomys und 22 % Muscardinus. Für jede Art wird die Verteilung auf zwei Rasterkarten (Rastergröße 25 km?) dargestellt, einmal nach der Fundzeit und dann nach der Fundzahl abgestuft. Außerdem werden Höhenverbreitung, Habitat und Exposition des Standorts geschildert. Typische Lebensräume werden in guten Fotos illustriert. Schließlich wird die Frage diskutiert, wie weit Konkurrenz zwischen den drei Arten ihre Verbreitung beeinflußt. Alle drei Schlafmäuse sind auf die Osthälfte Niedersachsens beschränkt. Ihre Nordgrenze ver- schiebt sich in der Reihenfolge Haselmaus, Siebenschläfer und Gartenschläfer nach Süden. Seit TEntUSs (1958) ist dies die erste umfassendere Darstellung der Verbreitung der Schläfer in Niedersachsen und die erste gründliche Kartierung überhaupt. Bemerkenswert ist besonders, daß die Haselmaus offenbar im ganzen westlichen Niedersachsen fehlt, ferner die Vorliebe des Gartenschläfers für Fichten in reinen und gemischten Beständen. Die Arbeit ist gründlich und auch in der Form ansprechend, doch wären eine noch ausführlichere Darstellung der neuen Befunde zu wünschen, wogegen manche Verallgemeinerung entbehrlich ist. So sollten die Aussetzungen des Siebenschläfers auf den Karten ebenfalls hervorgehoben werden. Neue Feststellungen sollten bezüglich ihrer Zuverlässigkeit (Samm- lungsexemplar, Nistkastenkontrolle oder bloße Sichtbeobachtung) behandelt werden, damit sie sich kritisch würdigen lassen. Zu den unzulässigen Verallgemeinerungen gehört, daß Haselmäuse immer in die Bodenvegetation flüchten (S. 24), denn nach ZıppeLıus und GoETHE (1951) klettern sie bei Störung stets aufwärts. Auch bauen sie nicht als einzige Schläfer freistehende Nester ($. 24), sondern auch Siebenschläfer (z. B. BAUER 1960) und Gartenschläfer (z. B. KAHumann und THoms 1974) können dies tun, und bei Bonn fand ich zumindest mehrfach vom Gartenschläfer überbaute und zum Teil als Wochenstube genutzte Amselnester. Im Rheintal schließen sich Sieben- und Gartenschläfer weniger wegen unter- schiedlicher Habitatnutzung aus (entgegen S$. 43). Vielmehr gibt es hier über weite Strecken rechts nur Sieben-, links nur Gartenschläfer bei ähnlichen Lebensräumen. Die Schlafmäuse sind in Okologie und Verhalten viel variabler, als regional begrenzte Befunde vermuten lassen. Gerade deshalb sind gründliche Untersuchungen wie die vorliegende wichtig und nützlich. J. NIETHAMMER, Bonn Kunrn, H.-].; ZELLER, U. (eds.): Morphogenesis of the Mammalian skull. Mammalıa depicta Vol. 13. Hamburg, Berlin: Paul Parey 1987. 144 pp., 68 fıgs., 5 tables. DM 80,-. ISBN 3-490-17718-5 Der vorliegende Sammelband enthält 7 Beiträge zur Morphogenese des Säugerschädels, die auf dem 7.Europäischen Anatomen-Kongref 1984 ın Innsbruck das Thema einer speziellen Veranstaltung bildeten. Sie sind in ihrer Gesamtheit ein eindrucksvolles Dokument für das neu erwachte Interesse an vergleichend-anatomischer Forschung und belegen die Ergebnisse, die in der Morphologie in den letzten Jahren erreicht wurden. Kunn bespricht in einer synthetischen Einleitung unter Berücksichtigung morphologischer, ontogenetischer, paläontologischer und funktioneller Gesichtspunkte alle jene Probleme, die sich aus dem Wandel vom Schädel der Reptilien zu dem der Säugetiere (Monotremata und Theria) ergeben (Kiefergelenks-/Gehörknöchelchenproblem, Natur und Herkunft der einzelnen Skelettelemente, Exo-Endoskelett, Sekundärknorpel, Ohrkapsel, Seitenwandbildung und Aufbau eines Syncraniums) in knapper und überzeugender Form. ZELLER behandelt als Beispiel für einen basalen Eutherier eine geschlossene Serie von 26 Embryonal- und einigen Postnatal-Stadien von Tupaza und ist damit in der Lage, ein differenziertes Bild vom Ablauf der Entwicklungsprozesse für den ganzen Ontogenese- ablauf zu entwerfen und eine Reihe offener Fragen zu klären. Besonders hervorgehoben seı der Beitrag (KUHN und ZELLER) über das Cavum epiptericum und die Seitenwandbildungen, dem ein reiches Material von Ornithorhynchus, Tachyglossus und Theria zugrunde liegt. Eine subtile Analyse der Inhaltsgebilde des Cavum epiptericum und des Aufbaus seiner Wand erbrachte eine Fülle neuer Befunde und eine Klärung alter Streitfragen. Nervenverlauf im Cavum und Beteiligung einer Lamina obturans an der Wandbildung sind beiden Monotremengattun- gen gemeinsam, doch ist Tachyglossus im Aufbau der sekundären Seitenwand höher organisiert als das Schnabeltier. Bei den Theria erfolgt die Reduktion der primären Seitenwand und der Aufbau einer sekundären Wand (Alisphenoid) in anderer Weise als bei den Monotremen. Beide Stammeslinien müssen außerordentlich früh eigene stammesgeschichtliche Wege eingeschlagen haben. Die Entwick- lungsvorgänge in der Orbitotemporalregion des Craniums zeigen in beiden Gruppen autapomorphen Charakter. Ahnlichkeiten im Endeffekt sind Konvergenzen. Die Arbeit bringt nıcht nur Klärung von 64 Buchbesprechungen Homologien, sondern enthält Hinweise auf funktionelle Zusammenhänge. Der Beitrag von W. MAIER bringt gleichfalls eine wichtige Erweiterung unserer Kenntnisse von der Morphogenese der Schädelsei- tenwand. Eine Reihe von 7 Stadien von Monodelphis domestica (Marsupialia, Didelphidae), ergänzt durch weitere Beuteltier-Arten, erweitert den Kenntnisstand wesentlich, auch in Hinsicht auf konstruktive und funktionelle Aspekte. Das Cavum epiptericum entsteht primär nicht als Folge der Hirnvergrößerung, denn es beherbergt bei Monodelphis in den Frühstadien ein riesiges Trigeminusganglion. Die Hirnentfaltung erfolgt erst in einer späteren Phase. SCHLIEMANN gibt anhand eines umfangreichen Materials von Entwicklungs- stadien fissipeder Carnivoren eine vergleichende Analyse der Knorpelstrukturen des Solum nasi, eines für die Phylogenie der Mammalıa wichtigen Merkmalskomplexes. Kııma berichtet über die Morpho- genese des Nasenskelettes der Zahnwale (41 Embryonalstadien, 6 Species) und führt für die meisten Strukturen des Chondrocraniums einen Vergleich mit der Nase terrestrischer Säuger zu einem überzeugenden Resultat. Der abschließende Beitrag von D. A. N. HoyTe behandelt das alte Problem der Beeinflussung der Schädelform und über diese Hirnform und Hirngröße durch Wirkung der Kaumuskulatur. Unter- sucht wurden mit Hilfe der Alızarin-Rot-Methode Wachstumsstadien von Kaninchen, Ratte, Meer- schweinchen und Schwein. Die Verteilung von Resorptions- und Appositionszonen konnte nicht mit der Wirkung der Muskeln eindeutig in Zusammenhang gebracht werden. Den ontogenetischen Formwandel vom kugligen zum gestreckten Neurocranium führt der Autor hingegen zum Teil auf Muskelwirkung zurück und vermutet auch eine Beeinflussung von Hirnform und Hirngröße. Die Ausstattung des Bandes ist von hervorragender Qualität. Die Gesamtheit der Beiträge bietet eine gute und koordinierte Übersicht über moderne Probleme der Craniologie und dürfte für Morphologen, Embryologen, Paläontologen und Systematiker unentbehrlich sein. D. STARcK, Frankfurt/M. BEGoN, M.; HARPER, J. L.; TownsenD, C. R.: Ecology. Oxford: Blackwell Scientific Publications 1986. 888 pp-, 476 ill. £ 14,50. ISBN 0-632-01339-7 Der Inhalt des Buches ist nur richtig einzuschätzen, wenn man den Untertitel kennt: Individuen, Populationen und Gemeinschaften. Die 3 Autoren mit ihren unterschiedlichen wissenschaftlichen Schwerpunkten brachten sowohl die Zoologie als auch die Botanik ein. Sie bauen auf den Eigenschaf- ten der Individuen auf, die im Lichte ihrer Eignung für das Überleben im Lebensraum gesehen werden. Die Ebene der Population wird schwerpunktartig berücksichtigt und die Struktur des Okosystems unter den Gesichtspunkten der Produktion, des Stofftransports und der Artenkoexistenz behandelt. Die beiden Kapitel „Prädation“ und „Ökosysteme“ beherrschen das Buch. Es zeichnet sich durch seine „Materialsammlung“, die Abbildungen, aus: In 22 Kapiteln bietet es 488 Abb. an; viele aus Originalarbeiten entnommen, andere geschickt entwickelt. Das Durcharbeiten des umgangreichen Bandes wird erleichtert durch „Randmarken“, die gleichsam eine Annotierung darstellen. Der Insider findet sich leichter zurecht, für den Studenten bieten sie eine Möglichkeit, sein Wissen autodidaktisch zu überprüfen. Die Randmarkierungen sind außerdem auch deshalb notwendig, weil das Inhaltsver- zeichnis zu wenig detailliert ausgefallen ist. Mit rund 900 Zitaten bildet das Literaturverzeichnis eine fundierte Grundlage. Den Umfang verdankt das Lehrbuch seiner Zielsetzung, nämlich sowohl theoretische Grundlagen zu bieten als auch die Ergebnisse von Felduntersuchungen darzustellen und anschließend in einer breiten kritischen Diskussion zu beleuchten. Wegen seines Aufbaus und seiner Aktualität kann das Lehrbuch sowohl für die Lehre als auch für das Studium uneingeschränkt empfohlen werden. R. SCHÖPFER, Osnabrück Resös, J.: Die grüne Hölle - ein bedrohtes Paradies. Bericht aus dem Regenwald. Hamburg und Berlin: Paul Parey 1987. 130 S., 158 Abb. DM 39,80. ISBN 3-490-23018-3 Das vorliegende Buch ist das Ergebnis mehrerer Reisen des Verfassers in die Urwaldgebiete Mittel- und Südamerikas. In einer mit unzähligen biologischen Fakten angereicherten, gut bebilderten Reiseschilderung berichtet er über die ungeheure Formenvielfalt und Komplexität der Lebensgemein- schaft Regenurwald, bevor er im 2. Teil des Buches auf die aktuelle Bedrohung des Okosystems Regenwald durch den Menschen und auf die Auswirkungen der Zerstörung dieses Okosystems auf den Gesamtwasserhaushalt und die Großklimalage eingeht. Die Waldzerstörung hat zwei Ursachen: Die Bevölkerungsexplosion und großflächige Rodungen durch internationale Konzerne. Als Alterna- tive für eine umweltschonende Regenwaldnutzung empfiehlt der Verfasser hier neben dem sinnvollen Einsatz von Kunstdünger einen Saatwechsel und Multistrata-Produktionssysteme. |]. LANGE, Berlin Deutsche Gesellschaft für Saugetierkunde 60. Hauptversammlung in Stuttgart, 28.September bis 2. Oktober 1986 Kurzfassungen der Vorträge und Posterdemonstrationen. Herausgegeben von Fritz Dieterlen, Stuttgart. Zusammenstellung: Pia Wilhelm, Stuttgart. 1986. 99 Seiten. Kartoniert 24,- DM Ein Hauptziel der Deutschen Gesell- schaft für Säugetierkunde ist, auf ihren Jahrestagungen über Säugetiere arbeitende Wissenschaftler verschie- denster Fachrichtungen zusammen- zuführen, den Gedanken- und Erfah- rungsaustausch anzuregen, um so Erkenntnisse aus den einzelnen Forschungsgebieten zu integrieren. Da aufeiner Jahresversammlung nicht alle Fachrichtungen gleichermaßen vertreten sein können, wurden - wie schon in den Vorjahren - wieder Schwerpunkte gesetzt: Erstens, die Paläontologie und, damit verbunden, Fragen der Systematik und Evolution der Säugetiere. Zweitens, ein Themenkreis, der vor- wiegend physiologische und etholo- gische Probleme bei Säugetieren um- faßt, und drittens, ökologische Themen, besonders zu Fragen des Artenschutzes und der Wildbiologie. Sehr erfreulich ist, daß eine größere An- zahl von Beiträgen den Fledermäusen gewidmet ist- der bei uns am stärksten bedrohten Säugetierordnung. Zu beziehen durch jede Buchhandlun Verlagsbuchhandlung Paul Parey Spitalerstraße 12 2000 Hamburg 1 PAUL P/REY Erscheinungsweise und Bezugspreis 1988: 6 Hefte bilden einen Band. Jahresabonnement Inland: 308,- DM zuzüglich 13,80 DM Versandkosten, Gesamtpreis 321,80 DM einschließlich 7% Mehrwert- steuer. Jahresabonnement Ausland: 308,- DM zuzüglich 18,- DM Versandkosten. Das Abonne- ment wird zum Jahresanfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, Spitalerstraße 12, D-2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesellschaft für Säugetier- kunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. Z. Säugetierkunde 53 (1988) 1, 1-64 Umwelt- und Naturschutz R sind machbar, Herr Nachbar! pektakuläre Umweltkatastrophen, wie sie sich in Tschernobyl und beim Schweizer Chemiekonzern Sandoz ereignet haben, machen es auch Gutwilligen schwer, an reale Fortschritte im Bereich Umwelt- und Naturschutz zu glauben. Aber es sind nicht nur die großen Katastrophen, die unsere Lebensgrundlagen bedrohen. Die öko- logische Zeitbombe besteht aus einem Mosaik großer und kleiner Katastrophen vor der | Haustür: Flurbereinigung, Begradigung von Gewässern, Straßenbau, Bodenvergiftung, Trinkwasserverseuchung, Luftverschmutzung, Fischsterben, Schadstoffeinleitungen in Flüsse etc. Diesen bedrohlichen Zustand zu entschärfen, Wege und Lösungen zur Besserung aufzuzeigen, ist das Anliegen dieses Buches. Es analysiert die gegenwärtige Natur- und Umweltkrise anhand von Schwerpunktbereichen (Gewässer, Moore, Dörfer und Städte, Straßenbau, Land- und Forstwirtschaft sowie Fremdenverkehr) und erläutert zahlreiche Lösungsmöglichkeiten. Endlich ein Buch zu diesem brennenden Thema, das zum praktischen Handeln anregt und jeden von uns auffordert: »Umwelt-und. Naturschutz sind machbar, Herr Nachbar!« Zu beziehen durch jede Buchhandlung. %* Wolf-Eberhard Barth: Praktischer Umwelt- und Naturschutz. Anregungen für Jäger und Forstleute, Landwirte, Städte- und Wasserbauer sowie alle anderen, die helfen wollen. 1987. 310 Seiten und 16 Tafeln. Mit 69 Fotos, davon 33 farbi 82 Zeichnungen und 27 Tabellen. Kartoniert 48,- DM Verlag Paul Parey Hamburg und Berlin PAUL PREV Da wa AL 12),65-128, April 1988 ISSN 0044-3468 C 21274 F ZITSCHRIFT FÜR | SAUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Organ der Deutschen Gesellschaft für Säugetierkunde Ahlen, I.: Sonar used by flying Lesser horseshoe bats, Rhinolophus hipposideros (Bechstein, 1800) (Rhinolophidae, Chiroptera), in hunting habitats. — Ortungslaute von fliegenden Kleinen Hufeisennasen, Rhinolophus hipposide- ros (Bechstein, 1800) (Rhinolophidae, Chiroptera), in Jagdbiotopen 65 Balakrishnan, M.: Structure of Lepus nigricollis hair from various body regions with Scanning Electron Microscopy. — Struktur der Haare von Lepus nigricollis aus verschiedenen Körperregionen im rasterelektronenmikroskopischen Bild 69 Storch, G.: Eine jungpleistozäne/altholozäne Nager-Abfolge von Antalya, SW-Anatolien (Mammalia, Rodentia). — An upper Pleistocene/lower Holocene rodent succession from Antalya, SW Anatolia (Mammalia, Rodentia) 76 3aker, Carolyn M.: Vocalizations of captive Water mongooses, Atilax paludinosus. — Vokalisationen des Sumpf- ichneumon, Atilax paludinosus, in Gefangenschaft 83 Mlikovsky, J.: Secondary sex ratio in the Przewalski horse Equus przewalskii (Mammalia: Equidae). — Das sekundäre Geschlechterverhältnis beim Przewalski-Pferd Equus przewalskii (Mammalia: Equidae) 92 Herzog, S.: The karyotype of the European roe deer (Capreolus capreolus L.). -— Der Karyotyp des europäischen Rehes (Capreolus capreolus L.) 102 Peters, J.: Osteomorphological features of the appendicular skeleton of African buffalo, Syncerus caffer (Sparrman, 1779) and of domestic cattle, Bos primigenius f. taurus Bojanus, 1827. - Osteomorphologische Unterscheidungs- merkmale am Gliedmaßenskelett vom afrikanischen Büffel (Syncerus caffer) und vom Hausrind (Bos primigenius f. taurus) 108 Wissenschaftliche Kurzmitteilungen Dort, Madeleine van: Note on the skull size in the two sympatric Mouse Deer species, Tragulus javanicus (Osbek, 1765) and Tragulus napu (F. Cuvier, 1822). — Bemerkung über die Schädelgröße von zwei sympatrischen Hirschferkel-Arten, Tragulus javanicus (Osbek, 1765) und Tragulus napu (F. Cuvier, 1822) 124 Kurre, J.; Fuchs, E.: Nachtaktivität von Spitzhörnchen (Tupaia belangeri). — Night activity of Tree shrews (Tupaia belangeri) 126 ANTHSONIAN” JUN 2 8 1988 Buchbesprechungen LIBRARIES Verlag Paul Parey Hamburg und Berlin HERAUSGEBER/EDITTORS P. J. H. van BREE, Amsterdam - W. FIEDLER, Wien - H. Frıck, München - W. HERRE, Kiel - H.-G. Krös, Berlin - H.-J. Kunn, Göttingen - E. KuLzer, Tübingen - B. Lanza, Florenz — J. NIETHAMMER, Bonn - H. REIcHSTEIn, Kiel - M. Röhnrs, Hannover - D. STARcK, Frankfurt a. M. - F. Strauss, Bern — E. THENIUS, Wien - P. VoceL, Lausanne SCHRIETEEITUNG/EDITORTATSOFEICE H. SCHLIEMANN, Hamburg - D. Kruska, Kiel This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con- tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information Die Zeitschrift für Säugetierkunde veröffentlicht Originalarbeiten und wissenschaftliche Kurzmittei- lungen aus dem Gesamtgebiet der Säugetierkunde, Besprechungen der wichtigsten internationalen Literatur sowie die Bekanntmachungen der Deutschen Gesellschaft für Säugetierkunde. Verantwort- licher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Harald Schliemann. Zusätzlich erscheint einmal im Jahr ein Heft mit den Abstracts der Vorträge, die auf der jeweiligen Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde gehalten werden. Sie werden als Supplement dem betreffenden Jahrgang der Zeitschrift zugeordnet. Verantwortlich für ihren Inhalt aa ausschließlich die Autoren der Abstracts. Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z. Hd. Prof. Dr. Dieter Kruska, Institut für Haustierkunde, Biologie-Zentrum, Neue Universität, Olshausenstr. 40-60, D-2300 Kiel. Für die Publikation vorgesehene Manuskripte sollen gemäß den „Redaktionellen Richtlinien“ abgefaßt werden. In ihnen finden sich weitere Hinweise zur Annahme von Manu- skripten, Bedingungen für die Veröffentlichung und die Drucklegung, ferner Richtlinien für die Abfassung eines Abstracts und eine Korrekturzeichentabelle. Die Richtlinien sind auf Anfrage bei der Schriftleitung und dem Verlag erhältlich. Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der Korrekturfahnen erfolgen. Vorbehalt aller Rechte: Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei- ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grund- sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber- rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of the first page of an article in this journal indicates the copyright owner’s consent that copies of the article may be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective, or for resale. For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the ECe®: Mit einer Beilage des Verlages Paul Parey Fortsetzung 3. Umschlagseite © 1988 Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstr. 12, D-2000 Hamburg 1; Lindenstr. 44-47, D-1000 Berlin 61. - Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., D-2240 Heide/Holst. Z. Säugetierkunde 53 (1988) 65-68 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Sonar used by flying Lesser horseshoe bats, Rhinolophus hipposideros (Bechstein, 1800) (Rhinolophidae, Chiroptera), in hunting habitats By I. AHLEn Department of Wildhfe Ecology, Swedish University of Agricultural Sciences, Uppsala Receipt of Ms. 21. 1. 1987 Abstract Flying lesser horseshoe bats were recorded in their natural hunting habitats at a number of different localities in Spain. The common sonar pulse consisted of about a 50 ms long CF-signal with a short upward sweep at the start and a downward sweep at the end. The CF-portion of the pulse had its strongest component at about 110 kHz. This is the second harmonic while there is a weaker first harmonic at about 55 kHz. The pulses were repeated about ten times per second. The material did not show much varıation which suggests that the auditory system ıs sharply tuned to the optimal frequency and that the local populations do not show much acoustic variation. One observation indicated that the lesser horseshoe bat might hunt from a perch like a flycatcher. Introduction Since the studies of ultrasonic emissions of the lesser horseshoe bat by Kay and PickvAanceE (1963) the sonar of this species has been known only in general terms. Good sonar data from hunting situations ın natural habitats have not been published. From other species we know that there may be important differences between sonar used indoors and outdoors; in outdoor flight the pulses are more developed in shape and specificity than in indoor flight (Arten 1981). Is the small varıiation ın frequency described by Kay and PıckvAncE (1963) for bats flying indoors still valid in a variety of natural field situations? Lesser horseshoe bats are considered to have the highest frequency of all European bats, but are the constant frequency signals pure tones or are there harmonics? Material and methods The ultrasonic sonar emitted by lesser horseshoe bats, Rhinolophus hipposideros was recorded on a number of different geographical localities in Spain 1982, 1985 and 1986. The bats were recorded when found in the hunting habıtats in the surroundings of their roosts. The recordings were 1982 made with a prototype to the D-920 frequency dividing detector (AHLEN et al. 1984) and a cassette recorder. In 1985 recordings were made with a further developed version of the same detector, called D-940. In addition to using cassette recorders, high frequency signals were also recorded with a Racal Store 4 D instrumentation tape recorder. In 1986 most recordings were done with the detector version D-960 which contains a “time expansion’ unit (signals stored in a digital memory and read to the recorder in one tenth of the original speed) (PETTERSSON 1986). In 1986 visual observations were made with the aid of an image amplifier, Wild Heerbrugg Big2. Analyses were made of a selection of recordings from Sevilla (Andalusia), Leon (Leon), Oviedo (Asturias), Huesca (Aragon) and Barcelona (Catalonıa) provinces (regions) of Spain. The signals were inspected and measured by the use of a digital memory oscilloscope, an FFT-analyser and by making sonagrams. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5302-0065 $ 02.50/0 66 I. Ahlen (©) 1 2 = Fig. 1. A pulse train from a lesser horseshoe bat Rh. hipposideros flying past the observer shown as an oscillogram (relative amplitude against time) 150 100 Fıg. 2. A single sonar pulse shown as a sonagram with frequency against time. A weak first harmonic at 54 kHz and a strong second harmonic at 108 kHz Results The lesser horseshoe bats were found hunting along hillsides or steep clıtfs, in small open spaces in scrubland and along low tree galleries at the edge of streams and ponds. Flying bats were also observed along walls of big buildings and stonedikes. Its sonar could only be heard at a short distance, with the D-940 (and D-960) only about 10 meters or even less. The most common sonar type used ın these environments had remarkably small variation and was easy to distinguish from the two other Rhinolophus species (ferrum-equinum and euryale) occuring in the same areas. The sonar signals consisted of a fast pulse train with about 50 ms long pulses (40-69 ms, average 48 ms, n = 21) repeated regularly with a little less than 100 ms between the start of each pulse (64-99 ms, average 89 ms, n = 19) (Fig. 1). Each pulse consisted of a long constant frequency signal with the strongest component at about 110 kHz (106-111, average 109 kHz, n = 23), a short upward sweep at the startand a corresponding downward sweep at the end stopping at about 90 kHz. The analyses revealed that there is a weaker first harmonic at about 55 kHz, which means that it is the second harmonic which has the most energy (Figs. 2-3). After this discovery I have noticed that the fundamental ıs easıly heard in the field by tuning the heterodyning to 55 kHz. Some analysed pulses also showed a faint third harmonic at about 165 kHz (Fig. 3). The occurrence of weak fundamentals together with strong second harmonics have been Sonar used by Lesser horseshoe bats in hunting habitats 67 =®) 100 105: © kHZ Fıg. 3. FFT-analysıs (Fast Fourier Transform, with Hanning weighting) of a CF-part of a pulse showing a maximum sound pressure at 110 kHz for the second harmonic (B), a weaker peak at 55 kHz (-42 dB relative to B) for the fundamental (A) and a third harmonic at 165 kHz (C) (-46 dB relative to B) discussed e.g. by SaLEs and PyeE (1974, p. 58) and was described for Pteronotus parnelli (SuGA 1984). The lesser horseshoe bats were also using other sounds at times, especially when they were circling around the entrance to their roosts or when they were hanging on twigs or small rocky outcrops. In the latter cases the differences mainly consisted of a varıied pulse length and repetition rate. My material is still insufficient to give a detailed description of these sound types. At one occasion I made an observation suggesting that Rh. hipposideros can use the “flycatcher’ behaviour described in tropical Rhinolophus species (SCHNITZLER et al. 1985). A lesser horseshoe bat was observed hanging on a small rocky outcrop. After a while it flew away out in the vegetation. I could hear it fly around but lost contact with it very soon. Coming back to the rock a couple of minutes later, I found the bat hanging on exactly the same place again. Discussion Bats with CF-components can separate their frequencies individually and thus might avoid interference (MILLER and DEcn 1981). They are likely to return to their optimal frequen- cies when hunting alone. Rhinolophid bats can compensate for doppler shifts to keep echo within a narrow band of best auditory frequencies (SCHNITZLER and HEnson 1979; SALES and PyE 1974). The ultrasonic sounds used by Rhinolophus hipposideros in a summer nursery colony (indoors) were studied in England by Kay and Pıckvance (1963). They reported that the female bats had a very small range of frequencies, only 3 kHz, from 110-114 kHz. The small range of frequencies seems surprising since about 60 bats were present. K.-G. HELLER (pers. com.) recorded hand-held specimens of five Rhinolophus-species where Rh. 68 I. Ahlen hipposideros of varıous orıgin (Southern Germany and Greece) ranged from 105 to (da: When comparing the same bat species from a number of geographically different localities it could not be excluded that there ıs a greater varıation in frequency. Therefore it is noteworthy that even my data on Rh. hipposideros from a number of colonies in different parts of Spain did not show much variation. This suggests that the auditory system is sharply tuned to the optimal frequency of the species and that the local populations do not show much acoustic variatıon. Acknowledgements This study was made possible in 1982 in connection with a study travel financed by the Swedish National Environmental Protection Board, in 1985 with the Spanish state’s grants for scientific cooperation between Spain and Sweden and in 1986 with economical support from the Swedish University of Agrıcultural Sciences. I thank Dr. L. MırLLer, Odense, for valuable suggestions when analysing the material and preparing an earlier draft of this article. I also thank Dr. H. BAAGoE, Copenhagen, for cooperation in the field work and for crıtical comments on the manuscript. Finally I thank Dr. K.-G. HELLER, Erlangen-Nürnberg, for kindly providing me with results from his sound recordings. Zusammenfassung Ortungslaute von fliegenden Kleinen Hufeisennasen, Rhinolophus hipposideros (Bechstein, 1800) (Rhinolophidae, Chiroptera), in Jagdbiotopen Laute der fliegenden Kleinen Hufeisennasen wurden mit Tonbandgerät in natürlichen Jagdbiotopen in vielen Lokalitäten ın Spanien aufgenommen. Der gewöhnlichste Ortungslaut besteht aus einem etwa 50 Millisekunden langen konstantfrequenten Signal mit einem kurzen frequenzmodulierten Anfangs- und Endteil. Der Konstantfrequenzteil hat die stärkste Komponente mit etwa 110 kHz. Das ist der erste Oberton, während der Grundton beı 55 kHz schwächer ist. Die Laute werden etwa zehnmal pro Sekunde ausgesendet. Die Variation an Ortungslauten zwischen Individuen und Populationen war sehr klein, wahrscheinlich ein Ausdruck für einen sehr engen reizbaren Frequenzbereich im Gehöror- gan. Eine Observation deutet an, daß Kleine Hufeisennasen die Jagdtechnik der Fliegenschnäpper benutzen. Literature AHL£n, I. (1981): Identification of Scandinavian bats by their sounds. Sw. Univ. Agr. Scı., Dept. of Wildlife Ecology, Uppsala, Sweden, Rep. 6, 1-56. AHLEN, I.; PETTERSSON, L.; SVÄRDSTRÖM, A. (1984): An instrument for detecting bat and insect sounds. Myotis 21-22, 82-88. Kay, L.; Pıckvance, T. J. (1963): Ultrasonic emissions of the Lesser Horseshoe Bat Rhinolophus hipposideros (Bech.). Zool. Soc., London, Proc. 141, 163-171. MILLER, L. A.; Decn, H. J. (1981): The acoustic behavior of four Vespertilionid bats studied in the field. J. Comp. Physiol. 142, 67-74. PETTERSSoN, L. (1986): Instruments for detection and analysis of ultrasonic signals. Inst. Technology, Uppsala university. Rep. UPTEC 86126R. SaLes, G.; PvE, D. (1974): Ultrasonic communications by animals. London: Chapman and Hall. SCHNITZLER, H.-U.; HACKBARTH, H.; HEILMANN, U.; HERBERT, H. (1985): Echolocation behavior of rufous horseshoe bats hunting insects in the flycatcher-style. J. Comp. Physiol. A 157, 39-46. SCHNITZLER, H.-U.; Henson, ©. W. (1980): Performance of airborne anımal sonar systems: I. Microchiroptera. In: Animal Sonar Systems. Eds. R.-G. Busner and J. F. Fısu. New York and London: Plenum Press. pp. 109-181. Suca, N. (1984): Neural mechanisms of complex-sound processing for echolocation. Trends NeuroScı. 7, 20-27. Author’s address: Prof. Dr. INGEMAR AHL£n, Department of Wildlife Ecology, Swedish University of Agricultural Sciences, Box 7002, S-75007 Uppsala, Sweden Z. Säugetierkunde 53 (1988) 69-75 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Structure of Lepus nigricollis hair from various body regions with Scanning Electron Microscopy By M. BALAKRISHNAN Department of Zoology, University of Kerala, Trivandrum, India IReceipt 05 Ms21,1.,1987% Abstract Scanning Electron Microscopic (SEM) studies were made on the morphology of hair samples of the head, neck, dorsal body, ventral body, fore-lımb, hind-limb and tail regions of the Indian hare, Zepus nigricollis. The hairs of varıous body regions tend to differ in colour and size. The type and pattern of arrangement of cuticular scales of these hairs also vary considerably from region to region. Scanning electron microscopic studies on the morphology of hairs of various body regions of any mammalıan species form a useful tool in the identification of prey species by the analyses of faecal matter of predators, and for evidences for the presence of various species of animals in any forest habitat. Introduction “Trichology’, the scientific study of haırs has specific relevance in the dietary investigation of carnıvores and field survey of mammals (Day 1966; DREYER 1966; RyDEr 1973; PERRIN and CAMPBELL 1979; KEoGH 1983; Buys and KEocH 1984). Data obtained from recent wildlife surveys in various forests of Kerala make it possible to determine that there ıs a need for the identification of varıous mammalian hairs, especially those of the prey species, which facılitate the conclusive identification of the presence of various mammal hairs in the faeces of predators (VıJayan et al. 1979; BALAKRISHNAN 1984; BALAKRISHNAN and EAsA 1986). The faecal matter of most of the wild predators have considerable quanta of hair and occasionally have undigested bones. The hairs of larger prey such as the sambar deer, Cervus unicolor and the spotted deer, Axis axıs; can be comparatively easily identified in the droppings of larger carnıvores such as the tiger, Panthera tigris and the wild dog, Cuon alpınus. However, a thorough identification of hairs commonly seen in the droppings of smaller predators such as the jungle cat, Felis chaus; the Asıatic jackal, Canıs aureus and the Indian fox, Vulpes bengalensıis, are ditficult. Hence, an attempt has been made to ıdentify hairs of a number of mammalıan species from varıous forests of Kerala with the aid of Scanning Electron Microscopy. The present report ıncorporates the data on the fine structure of hairs of various body regions of the Indian hare, Lepus nigricollıs. Material and methods Lepus nigricollis (Cuvier) of both sexes (four males and four females) were trapped from forest habitats in Wynad and Trivandrum, Kerala and hair samples were plucked carefully from their head, neck, dorsal body, ventral body, fore-lımb, hind-limb and tail regions using a fine forceps. These samples were kept in hexane or in 70 % alcohol for 1-72 h for cleanıng and were dehydrated in ascending grades of alcohol. Hair samples were cross sectioned at about the middle portion using a fine stainless steel knife. A few hairs were also sectioned longitudinally. The samples were then mounted on studs, dried using a vacuum dryer and gold coated ın a Type JEE 4B Vacuum Evaporator at high vacuum. These samples were scanned under a JOEL JEM 100C/ JOEL ]JSM 35 Scanning Electron Microscope at an accelerating voltage of 10 kV and studied at magnifications ranging from X300 to X10,000. For comparison, exposures from middle portions of the samples were used. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5302-0069 $ 02.50/0 709 M. Balakrishnan Results Observations on the morphology of haır collected from different regions did not reveal any marked sexual dimorphism and hence the data from male and female hares were combined. The table shows the data on size and pattern of colouration of hair samples of various body regions. In general, these hairs have a length of 10-25 mm depending on the body site of its Table. Showing the size and colour pattern of hairs of various body regions of the Indian hare, Lepus nigricollis Body region Hair size, mm* Colour pattern of hair Minımum Maximum Proximal Middle Distal Distal tip Head White-brown Black 7 GreamsBlack Neck Cream Grey Cream Black Dorsal body White Black 7 "Cream Black Ventral body White White White White Fore-lımb White White Cream Cream Hind-limb White Grey Cream Black Tail White Black Cream Black * Data represent a minimum of 30 samples from each region origin. The differential colour pattern also helps to identify hairs of one region from those of the other. The hairs of ventral body regions are particularly white and smooth, whereas those of other regions have bands of two to four colours such as white, black, cream and grey. The type and pattern of arrangements of cuticular scales of hairs of different body regions observed with the SEM revealed the following: Head and neck: The cuticular scales of the hairs of head and neck of the hare are flattened, conical-shaped and are tightly packed. There are a number of ridges and grooves on the surface of neck hairs in the longitudinal plane as a result of the pattern of arrangement of the scales (Fig. 1). The pattern of arrangement of the scales could easily be used to distinguish head hair from body regions, but neck hair and fore-lımb hair could not be distinguished by scale patterns. Dorsal body: The cuticular scales of dorsal body hair are flattened and are thickly packed (Fig. 2). Roots of these hairs are thick showing the typical patterns of roots of hare hairs (Fig. 3). Ventral body: The scales of ventral body haır are elongated in shape and are tightly packed. Elongated grooves on the cuticular surface are also seen as in the case of neck hairs. These hairs are deep-rooted. However, the root is thin (Fig. 4) when compared to the other hair roots. The medulla of these hairs is divided into two columns by the mid- medullary growth of cortical tissues (Fig. 5). Fore-limb: The scales of the hair of fore-lımbs are flattened and elongated, and the distal ends are conical-shaped (Fig. 6). These hairs are not deep-rooted. The proximal portions of them are more or less oval in shape. Cuticular scales are also seen on the surface of these roots (Fig. 7). Well developed medullary cells of these hairs are shown in Fig. 8. Hind-limb: The scales of the hair of hind-limbs are flattened and thickly packed along the entire length (Fig. 9). The cortex is thin and the medulla ıs extended with larger cells (Fig. 10). Tail: The cuticular scales of the hair of the tail are flattened and are similar to scales of hair of the dorsal body region. These scales are also tightly packed ın a uniform pattern er, ID) Fine structure of hare hair 71 Figs. 1-4. 1: The flattened, elongated and conical shaped cuticular scales of the head and neck haır. (x1,370); 2: Dorsal body hair showing the tightly packed pattern of arrangement of cuticular scales. (x1,370); 3: The typical structure of root of hairs of head and dorsal body. (x910); 4: The morphology of a root of ventral body hair. (Note the difference in the size and shape of roots of dorsal body hairs). (x 1,830) 72: M. Balakrishnan Figs. 5-8. 5: The cross section of a ventral body hair showing the partition of medulla into two columns. (X4,120); 6: Fore-limb hairs have elongated cuticular scales with conical shaped apices. (x1,280); 7: The oval shaped roots of fore-lımb hairs are shown. Note the development of cuticular scales on roots and the proximal openings of the medulla. (X 2,390); 8: The medullary cells of the fore- lımb hair showing the pattern of arrangement of chamber-like formation with the development of intercellular septa. (X 4,120) Discussion It has been emphasized that identification of ingested hair is a useful tool determining the feeding habits and food preferences of predators (BRUNNER and CoMAN 1974; PERRIN and CAMPBELL 1979). Identification of hairs may also play a salient role in the field of forensic science (ADORGAN and KoLENOoskY 1969; KEOGH 1983). However, only a few mammals- have been subjected to systematic studies on the structure of hair (NoBack 1951; LynE and McManon 1951; BENEDICT 1957; Staıns 1958; KHEMELEVASKAYA 1965; Day 1966; AporjJan and KoLEnoskY 1969; TREVOR-DEUTCH 1970; RyDER 1973; BRUNNER and CoMan 1974; KoPpPIker and Sasnıs 1976, 1977; PERRIN and CAMPBELL 1979; SOKOLOV 1979; KEoGH 1983; RAaJarAM and MENoN 1986). Despite the structural differences of cuticular scales of different species of mammals discernible during these studies Fine structure of hare hair 3) IE N N ir N N RÜ \ S > zw Tr Sa Fıigs. 9-11. 9: Showing the hind-limb hairs with tightly packed flattened scales. (x 1,540); 70: A cross section of a hind-limb hair showing the pattern of arrangement of medullary cells. (<2,340); 71: The tail hair with tightly packed, flattened cuticular scales. (X 1,180) 74 M. Balakrishnan (HAsHIMOTO and SHIBAZART 1975), major distinguishing features of hairs of different body parts of the same animal were not observed. The present study made it possible to observe distinguishing features in the types, and in the pattern of arrangement of cuticular scales of hairs of various body regions of L. nigricollis. The type and pattern of arrangement of the cuticular scales of hair are found to be prominent factors in differentiating the hair of one region from those of the other. In addition to the patterns of arrangement of cuticular scales of the cortex, the medullary cells also help to differentiate the hair of various body regions. These features can be clearly observed by scannıng diverse samples of haır of different body regions, and can depend on these for the identification of hair samples of various species (RAJARAM and MENoN 1986) as well as hair of different regions of the same species. Faecal analyses of predators indicate that the droppings have lots of hairs of prey anımals. Undigested materials in the faeces of predators would provide information on food preferences of any anımal species. Hence identification of hair samples in the droppings of predators would be helpful for studies on the feeding habits of predators. This necessitates a thorough investigation, on the structure of hair samples of varıous body region of all the prey anımals in any forest. Scanning electron microscopic studies on the morphology of the hairs form a comparatively easy and fool-proof technique to identity the hairs of any species. Since the fine surface structure of hair of varıous body regions of the same anımal differ considerably, it is necessary to study the hair samples of various body regions of any species for a knowledge on the structure of body hairs and for detailed analyses of faecal remains of hair samples. Although many of the smaller prey anımals could not be seen directly during field surveys (BALAKRISHNAN 1984; BALAKRISHNAN and EAsAa 1986), the presence of hair in the faeces of predators provides indirect evidence for the presence of these species in the forest area. Further detailed studies on the fine structure of various mammalian hair would be helpful for identification of hair which may be seen on fences, twigs and on the habitat and exploratory grounds of anımals. This, ın turn would be helpful for species ıdentification and for inferences of anımal movements in any forest habıtat. Acknowledgements Part of this investigation was carried out at the A. N. Severtzov Institute of Evolutionary Anımal Morphology and Ecology of the USSR Academy of Sciences, Moscow during the tenure of an Academic Exchange Fellowship Programme of the Indian National Science Academy, New Delhi with the USSR Academy of Sciences, Moscow. I am grateful to Academician V. E. SOKOLOv and Dr. T. P. EvGENJEvA of the above Institute for facilities and for various helps rendered during the period of this investigation. I am also grateful to the Director of the CSIR Regional Research Laboratory, Trivandrum, Mr. PETER KosHy, Scientist-in-charge of the SEM and the technical staff of the Division for providing their facilities to carry out part of this investigation. I am thankful to Prof. K.M. ALEXANDER, University of Kerala for encouragement and for critically reading the manuscript. Financial assistance from the State Committee on Science, Technology and Environment, Govern- ment of Kerala ıs also acknowledged. Zusammenfassung Struktur der Haare von Lepus nigricollis aus verschiedenen Körperregionen im rasterelektronenmikroskopischen Bild Rasterelektronenmikroskopische Untersuchungen wurden an Haaren aus verschiedenen Körperregio- nen vom indischen Hasen Lepus nigricollis durchgeführt. Die Haare dieser Art zeigten regionale Unterschiede in Färbung und Größe sowie in der Ausprägung der Cuticular-Muster. Vergleichbare Untersuchungen an anderen Säugetieren können einen hilfreichen Beitrag zur Artbestimmung an Haaren liefern. Sie sind insbesondere dann von Nutzen, wenn das Beutetierspektrum von Raubtieren über Kotanalysen erfaßt werden soll. Fine structure of hare hair 75 References Aporjan, A. S.; KoLenosky, G. B. (1969): A manual for the identification of hairs of selected Ontario mammals. Res. Report (Wildlife) No. 90. Ontario: Dept. Lands and Forests. BALAKRISHNAN, M. (1984): The large mammals and their endangered habitats in the Silent Valley forests of South India. Biol. Conserv. 29, 277-286. BALAKRISHNAN, M.; Easa, P. S. (1986): Habitat preferences of the larger mammals in the Parambiku- lam Wildlife Sanctuary, Kerala, India. Biol. Conserv. 37, 191-200. Buys, D.; KeocH, H. ]. (1984): Notes on the microstructure of hair of the Orycteropodidae, Elephantidae, Equidae, Suidae and Giraffidae. S. Afr. J. Wildl. Res. 14, 111-119. BENEDICT, F. A. (1957): Hair structure as a generic character in bats. Univ. Calıf. Pub. Zool. 59, 285-548. BRUNNER, H.; CoMan, B. (1974): The identification of mammalıan hair. Melbourne: Inkata Press. Day, N. G. (1966): Identification of hair and feather remains in the gut and faeces of stoats and weasels. J. Zool., Lond., 48, 201-217. DREYER, J. H. (1966): A study of hair morphology in the family Bovidae. Onderstepoort J. Vet. Res. 1966, 379472. HasHIMOTO, K.; SHIBAZART, S. (1975): Ultrastructural study on differentiation and function of hair. In: Biology and disease of the hair. Ed. by K. Topa, Y. IsHIBAsHI1, Y. Horı and F. MoTIkAwa. Tokyo: Univ. Park Press. KEoc#H, H. J. (1983): A photographic reference system of the microstructure of the hair of southern Afrıcan bovids. S. Afr. J. Wildl. Res. 13, 89-132. KEHMELEVASKAYA, N. V. (1965): Structure of the rodent hair cuticle, its varıability and significance for taxonomy. Zool. Zhur. 40, 1064-1074. KoPPIkEr, B. R.; Sasnıs, J. H. (1986): Identification of hairs of some Indian mammals. J. Bombay naraElise. Soc. 73, 5 20; KoPPIkEr, B. R.; Sapnıs, J. H. (1977): Further studies on the ıdentification of hairs of some Indian mammals. J. Bombay nat. Hist. Soc. 74, 50-59. Lyne, A. G.; McManon, T. S. (1951): Observations on the surface structure of the hairs of Tasmanıan monotremes and marsupials. Pap. Roy. Soc. Tasmanıia 1950, 71-84. Noßack, C.R. (1951): Morphology and phylogeny of hair. Ann. N.Y. Acad. Scı. 53, 476492. PERRIN, M. R.; CAMPBELL, B. S. (1979): Key to the mammals of the Andries Vosloo Kudu Reserve (Eastern Cape), based on their hair morphology, for use in predator scat analysis. S. Afr. J. Wildl. Res. 10, 3-14. RAJARAM, A.; MEnon, R. K. (1986): A Scanning electron microscope study of the hair keratins of some anımals of the Indian sub-continent - A preliminary report. J. Bombay nat. Hist. Soc. 83, 427429. Ryper, M.L. (1973): Hair. London and Southampton: Institute of Biology. Camelot Press Ltd. SokoLov, V.E. (1979): Adaptive features of mammalıan hair. In: The adaptive features of epithelium and its derivatives — An atlas of scanning electron microscopic photographs. Moscow: Nauka publishing office. STaıns, H. J. (1958): Field key to guard hair of Middle Western furbearers. J. Wildl. Mgmt. 22, 95-97. TREVOR-DEUTCH, B. (1970): Hair morphology and its use in the identification of taxonomic groups. M. Sc. Thesis, Canada, McGill University. VıJayan, V. S.; BALAKRISHNAN, M.; EAsa, P. S. (1979): Periyar Tiger Reserve: a reconnaissance report. Peechi; Kerala Forest Res. Inst. Author’s address: Dr. M. BALAKRISHNAN, Department of Zoology, University of Kerala, Kariavat- tom 695 581, Trivandrum, Kerala, India Z. Säugetierkunde 53 (1988) 76-82 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Eine jungpleistozäne/altholozäne Nager-Abfolge von Antalya, SW-Anatolien (Mammalıa, Rodentia) Von G. STORCH Forschungsinstitut Senckenberg, Frankfurt am Main Eingang des Ms. 9. 2. 1987 Abstract An upper Pleistocene/lower Holocene rodent succession from Antalya, SW Anatolia (Mammalia, Rodentia) From Middle Paleolithic through Early Upper Paleolithic layers there was an increase of aridity and steppe conditions. While steppe predominated, minor ecological changes may have occurred during the “Epipaleolithic”. Obviously, Microtus gud was at least locally replaced in the Taurus Mts. by M. nivalıs, a rather recent western immigrant. Mus abbotti and Meriones tristrami made their first appearance in uppermost Pleistocene and Holocene layers, respectively. Some species did not survive the upper Pleistocene in the area under study. Nevertheless, a marked ecological and faunal stability becomes evident. Einleitung Die beschriebenen Nagerfunde stammen aus der Höhlenstation Karain, etwa 30 km NWN von Antalya an der anatolischen Südküste gelegen. Das Höhlensystem liegt am Rand der riesigen Travertinebene von Antalya in kreidezeitlichem Kalk des Taurus. Der Eingang in 450 m ü.NN überblickt die etwa 150 m tiefer liegende Ebene bis hin zum Meer. Die Ausgrabungen von Karain B fanden in Zusammenarbeit des Instituts für Urge- schichte der Universität Tübingen mit der Universität Ankara vom 16. Sept.—4. Nov. 1985 statt. Die pleistozänen Sedimente wurden mit einer Maschenweite von 1 mm geschlämmt und vor Ort auf Artefakte und Kleinfauna ausgelesen (ALBRECHT 1986). Daten zur jüngeren Kleinsäuger-Faunengeschichte und ihrem klimaökologischen Hin- tergrund in Anatolien sind außerordentlich dürftig. Das jungpleistozäne/altholozäne Profil von Karain B soll in dieser Hinsicht als ein Mosaikstein dienen. Ergebnisse und Diskussion Das jungpleistozäne/altholozäne Nagerprofil von Karain B Durchlaufende Arten sind Spalax nehringi, Apodemus mystacinus und Microtus guentheri. Gemeinsam belegen sie offenes, relatıy trockenes Gelände sowie nackte Gesteins- und Felspartien. Diese 3 Arten sind gleichermaßen kennzeichnend für die heutigen und die pleistozänen Kleinsäugerfaunen des westlichen Anatolien. Sie kommen im jüngeren Mittelpleistozän (?mittleres Steinheimium) von Chios vor (STORCH 1975; Chios war zu dieser Zeit Teil des anatolischen Festlands), Apodemus mystacinus und Spalax cf. nehringi liegen auch schon im Altpleistozän (Altbiharium) von Kalymnos vor (Kuss und STORCH 1978; auch die Dodekanes-Insel Kalymnos war zu dieser Zeit Festland). Die Nager der mittelpaläolithischen, über 30000 Jahre alten Abtragungseinheiten lassen im Ensemble auf relativ günstige Umweltverhältnisse schließen. Sciurus anomalus und Apodemus flavicollis benötigen Wald, zumindest aber Streifen dichter, höherer Vegetation. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5302-0076 $ 02.50/0 Zn. ‘sne Jyaru J97g93sdunyansıaruNn wr uonepasgqn ul “wAeruy “oyrurg T9q AstamydeN >fIssofqns usspıy9s y9opal “3524 Asny- A uUayosıjoreue 19P UoA Inu IST 299004 snwnuod "(gZ6L) HONOLIS pun (96T) NUOASO “S/61) IAAOTIanNny “T/61) ‘fe 9 NaL1A] ZI6L) SIAAOW pun 1aa80) “Z/61) Te 9 Aaydanasag yeu WWoNIoA 93nnad uezojof UrZOISIO]A wunytuor[eg wnytynpoyfeyY -ıdq | -3unf SI Y -PRIW 3unIonel] Jy9stdojorysIy (ds vjosıauy pnd sn1041 sııvaav snJ01 11agqguond SnI0A1 1uvasıı1 SOUOLAo N N SNIAILAOSI N SMAOIVAINU SNNIINAZ snunvisku snwopoddy syyosiavjf snwopody snarvaks snwopody 10ggP sn 1Bungou xojvds 199904 snunmokpy snpwour snanıDS snuudkıdogupx sSn]J21I JU9ZIy usNayurssdungengVy Dyın] eAjeyuy Joker ‘g urereyy uoA [J01d193eN Juezojoyype/suezogsispdäunf seq ‘I ayJogvL Eine jungpleistozäne/altholozäne Nager-Abfolge von Antalya, SW-Anatolien 78 G. Storch Für Microtus arvalıs und Arvicola sp. ist im ganzen Jahr verfügbare saftige krautige Vegetation und ausreichende Pflanzendeckung zu fordern. Microtus gud ıst an zerklüftetes Gestein gebunden und kommt auch im Wald vor, wo dieser nicht völlig geschlossen ist. Gegenüber der nahverwandten und ebentalls petrophilen Schneemaus, Microtus nıvalıs, ist M. gud stärker hygrophil (SPITZENBERGER 1971; STEINER 1972). Unter Einschluß der ım Protil durchlaufenden Offenlandbewohner und von Myomimus roachı, der heute steppenartige Lebensräume bevorzugt, zeichnet sich eine recht große ökologische Vielfalt ab. Im Älteren Jungpaläolithikum etablieren sich mit Citellus xanthoprymnus und Meso- cricetus brandti ausgesprochene Steppenbewohner, und auch das heutige Verbreitungs- gebiet von Cricetulus migratorins schließt zum größten Teil Steppen ein. Citellus und Cricetulus leben noch im Untersuchungsgebiet, nicht aber Mesocricetus brandtı, dessen Verbreitungsschwerpunkt heute im anatolischen Hochland liegt (SPITZENBERGER 1972). Die Zunahme von Bewohnern offener trockener Lebensräume geht auf Kosten von Arten, die hinsichtlich Feuchtigkeit und Vegetation anspruchsvoller sind. Apodemus flavicollis und Sciurus anomalus sind im Älteren Jungpaläolithikum nicht belegt, und die Nachweise von Apodemus sylvaticus, Arvicola sp. und Microtus arvalis enden nacheinander. Es zeichnet sich also eine leichte Verschiebung zu arıderen Verhältnissen ab. Arvicola verschwindet damit offenbar endgültig aus dem Untersuchungsgebiet, denn der rezente A. terrestris ıst in Anatolien nur von sehr wenigen Fundpunkten bekannt, die aber nicht den Südwesten einschließen (OsBorn 1962). Im „Epipaläolithikum“ müssen großräumig Steppenverhältnisse weiterbestanden haben (Citellus, Mesocricetus, Cricetulus). Ab der Abtragungseinheit 20 (C14 Datierung 14160 + 210 Jahre BP) könnte sich jedoch eine kurzzeitige Verbesserung lokalklimatischer Verhält- nisse widerspiegeln: Apodemus flavicollis und Microtus arvalıs sind nochmals nachzuwei- sen (Abtragungseinheit 19), und Sciurus tritt wieder auf (Abtragungseinheit 18). A. flavicollis und M. arvaliıs kommen im Untersuchungsgebiet nach diesem Zeitpunkt often- sichtlich nicht mehr vor (heutige Verbreitung von M. arvalıs s. FELTEN et al. 1971; für Apodemus fehlen allerdings neuere Daten). Gegen Ende der „epipaläolithischen“ Schich- tenfolge scheinen die klima-ökologischen Verhältnisse wieder etwas ungünstiger geworden zu sein. Das kontinuierliche Auftreten von Microtus gud über die jungpleistozäne Abfolge endet in Abtragungseinheit 17. Heute ist der nahverwandte M. nivalıs durch die gesamte Tauruskette bis in den Libanon und Antilibanon südwärts verbreitet, während M. gud nur von einem Reliktvorkommen aus dem östlichen Mitteltaurus bekannt ist (SPITZENBERGER 1971). Dieser Befund spricht dafür, daß M. nivalıs im Untersuchungsgebiet ein junger Immigrant von Westen her ist. Dort ist er auch schon im jüngeren Mittelpleistozän von Chios belegt (SrTorcH 1975). M. nivalıs ist weniger hygrophil, was für die zumindest gebietsweise erfolgte Verdrängung des alteingesessenen M. gud sicherlich von mitentschei- dender Bedeutung war. Mus abbotti erscheint im Karain-Profil erstmals gegen Ende der pleistozänen Schichtenfolge. Es ist eine offene und trockene Lebensräume bevorzugende Outdoor-Art der Hausmäuse. Zwischen der pleistozänen Serie und dem holozänen Schichtpaket ist ein Hıiatus von mehreren tausend Jahren anzunehmen. Es gibt, im Gegensatz zu jungpleistozänen Abtra- gungen, nur Einzelfunde aus dem Altholozän. Auffällig ist dabei, daß sie die frühesten Nachweise von Meriones tristrami einschließen (chalkolithische Abtragungseinheiten 10- 9). Dieser reine Trockensteppen-Bewohner könnte eine sich im älteren Holozän verschär- fende Aridität bezeugen. M. tristrami liegt schon im Altpleistozän SW-Anatoliens vor (Kuss und Storch 1978), so daß ihr Fehlen in der jungpleistozänen Abfolge nicht mit „mangelnder Verfügbarkeit“, sondern ungeeigneten Lebensbedingungen zu erklären wäre. Verglichen mit mitteleuropäischen Verhältnissen zeigt das jungpleistozäne/altholozäne Nagerprofil von Karain B erwartungsgemäß weit weniger betonte faunistische Verände- rungen. Unter Einschluß älterer Lokalfaunen (STORcH 1975; Kuss und STORCH 1978) wird Eine jungpleistozäne/altholozäne Nager-Abfolge von Antalya, SW-Anatolien LS überhaupt eine bemerkenswerte ökologische und faunistische Stabilität im Pleistozän SW- Anatoliens deutlich. Steppen hatten immer eine prägende Rolle gespielt, und nackte zerklüftete Gesteinspartien mußten immer verfügbar gewesen sein. Wälder und Feuchtge- biete unterlagen in ihrer Ausdehnung aber Schwankungen, wie sie auch in der Kleinsäuger- fauna des Karain-Profils reflektiert werden. Derzeit erscheint allerdings der Versuch, solche Oszillationen mit den in Mitteleuropa mehr oder weniger etablierten Biozonen parallelisieren zu wollen, wenig sinnvoll. Bemerkungen zu einzelnen Taxa Gattung Mus Mus abbotti Waterhouse, 1837 (= Mus spicilegus „Sud“ in ORrsınt et al. 1983): Die von KRATOCHVIL (1986) herausgearbeiteten Artkriterien an den M1/1 treffen im vorliegenden Material zu. Am MI/ sind t1 und t4 durch eine tiefe Mulde getrennt und treten an der lingualen Kronenkontur als selbständige Höcker hervor. Am M/1 finden sich vorn Protoconid, Metaconid, linguales und labiales Anteroconid als gut differenzierte Höcker. Unterschiede zu M. hortulanus am Jochbogen sind nicht überprüfbar, doch spricht die bedeutende Zahn- größe (M1/ = 1.84 x 1.00 mm) für M. abbottı. Gattung Apodemus Apodemus mystacinus (Danford und Alston, 1877): Die M1/ zeigen aus- nahmslos das für pleistozäne und re- zente anatolische Populationen kennzeichnende mystacinns-Muster (STORCH 1975, 1977; Kuss und STORCH 1978). T12 ist mit t9 und t8 verbunden, und die Verbindung t9-t12 ist zumindest so gut entwik- kelt wie die von t8-t9. Bei dem bal- 20 20 a) ad aaa a . 9 Abb. 1. Größte Längen (L) und Breiten (B) von MI/ der aseaen aan hingegen in Karain B vertretenen Muriden-Arten; alle Jungplei- steht a2 schlußeingulum-artig nur stozän. « Apodemus mystacinus, @ Apodemus flavicollis, mit t8 ın Verbindung. In der Konfi- O Apodemus sylvaticus, a Mus abbotti guration des t12 unterscheiden sich die beiden anderen Apodemus-Arten aus Karain B von A. mystacinus: A. flavicollis besitzt einen schlußcingulum-artigen t12, der an dem kleineren und zierlicheren M1/ von A. sylvaticus höchstens schwach angedeu- tet ist. A. mystacinus von Karaıin hat im Mittel etwas größere M1/ als heutige westanatolische Tiere, stimmt aber gut mit jung-mittelpleisto- zänem Material von Chios überein (STORCH 1975). Gattung Microtus | Abb. 2. M1/ von Karain B in Occlusalan- Microtus (Chionomys) gud Satunin, 1909: Die sicht. 1= Mus abbotti, 2 = Apodemus mysta- M3/ aus dem Jungpleistozän von KarainB _cinus 80 G. Storch Tabelle 2. Zahnmaße der drei Microtus-Arten von Karain B, SW-Anatolien Microtus gud Microtus guentheri Microtus arvalıs stimmen mit denen rezenter Tiere von dem einzig bekannten Fundpunkt im Taurus (bei Ulukisla ım östlichen Mitteltaurus; SPITZENBERGER 1971) überein. Beide besitzen ein normalis-Muster, d.h. 3 Labial- und 4 Lingualantiklinalen. Sie unterscheiden sich hierin von Populationen des heutigen ostpontisch-kaukasischen Areals der Art, die ausnahmslos komplexer gebaute M3/ besitzen (SPITZENBERGER 1971; STEINER 1972). Die M/1 stimmen dagegen überein. Der kennzeichnende Vorderlobus besteht aus einem konfluenten Dentin- feld und besitzt 2 Lingualantiklinalen und eine mehr oder weniger ausgeprägte labiale Kante. Es kommt nicht zur Abschnürung der Vorderkappe wie bei M. nivalıs oder zu einer völlig gerundeten Labialkontur des Vorderlobus wie bei M. oeconomus. Die rezente und die jungpleistozäne Population aus dem Taurus sind durch ihre M3/ gut charakterisiert, und sie sollten gegen die ostpontisch-kaukasischen Populationen subspezifisch abgegrenzt werden. Die Zuordnung isolierter M3/ zu den vertretenen Microtus-Arten erscheint mir relatıv unproblematisch. Die M3/ von M. guentheri sınd am kompliziertesten und besitzen charakteristische distale Enden (vgl. BESENECKER et al. 1972; STORcH 1975). Die M3/ von M. arvalıs sind einfach und weisen verrundete Distalenden auf (vgl. STORCH 1975). Darüber hinaus sınd die M3/ von M. arvalıs kleiner und zierlicher, was besonders in Seitenansicht auffällt. Es gibt keine zierlichen und gleichzeitig kompliziert gebauten M3/ als Hinweis auf M. socıalıs. Die M/1 von M. guentheri und M. arvalıs unterscheiden sich außer ın der Größe ın der von STORcCH (1975) dargelegten Weise. So besitzt M. guentheri ausgedehntere, aufgeblähte Vorderkappen mit häufiger Kantenbildung, während M. arvalıs relativ zierliche Vorder- kappen aufweist. (Es scheint unklar zu sein, ob in Kleinasien heute M. arvalıs oder M. epiroticus vorkommt. Die vorliegenden Zähne können nicht zu einer Entscheidung beitra- gen, und arvalıs ıst als konventioneller Gebrauch des Artnamens zu verstehen.) M. guentheri aus dem Jungpleistozän von Karain B weist deutlich größere M/1 auf als heutige westanatolische Tiere und als eine jung-mittelpleistozäne Population von Chios. Gute Übereinstimmung besteht jedoch zu altholozänem Material von Chios, das sich auch in seinen Skelettmaßen als sehr großwüchsig erweist (BESENECKER et al. 1972; STORCH 1975). Gattung Arvicola Arvıcola sp.: Das Schmelzband der M/1 ıst auf der konkaven und der konvexen Prismen- seite gleichdick („cantiana“-Muster). Dieses Stadium vermittelt zwischen der Konfigura- tion bei Mimomys savini und den stratigraphisch frühesten Arvicola-Arten (konvexes Schmelzband verstärkt) einerseits und dem Muster heutiger mitteleuropäischer A. terrestris (konkaves Schmelzband verstärkt) andererseits. Biostratigraphische Einstufungen anhand solcher Entwicklungen sind unter Umständen für einen Ausschnitt, nicht jedoch über das Gesamtareal der Gattung durchführbar. Die Schmelzbänder rezenter A. terrestris zeigen eine Kline vom typischen terrestris-Muster in W- und Mitteleuropa bis zur typischen „cantiana“-Form ın Vorderasien (RÖTTGER 1986). Das jungpleistozäne Material stimmt hinsichtlich der Schmelzbanddifferenzierung mit heutigen SW-asiatischen Populationen überein, und es ist darın fortschrittlicher als A. praeceptor aus dem jüngeren Mittelpleisto- Eine jungpleistozäne/altholozäne Nager-Abfolge von Antalya, SW-Anatolien 81 N... >93 Abb. 3. Arvicolidenzähne von Karain B in Occlusalansicht. Microtus gud: 3-6 = M/1, 7-9 = M3/; Microtus guentheri: 10-12 = M/1, 13-15 = M3/; Microtus arvalıs: 16-18 = M/1, 19-20 = M3/; Arvıcola sp.: 21-23 = M/1, 24-26 = M3/ zän von Chios (STORcH 1975). Der Artname wird zunächst offengelassen, bis besser erhaltenes Material die Überprüfung auch anderer Merkmale erlaubt. Zahnmaße von Arvicola sp. aus Karain: M/1 = 3.1-3.8, M3/ = 2.3-2.4 mm. Danksagungen Herr Dr. GERD ALBRECHT als örtlicher Grabungsleiter hat mir dankenswerterweise die Kleinsäuger- fauna überlassen. Die Zeichnungen fertigte Frau A. HELFRICHT, Senckenberg-Museum, an. 82 G. Storch Zusammenfassung Ausgehend von günstigeren Umweltverhältnissen im Mittelpaläolithikum nahmen im Älteren Jung- paläolithikum Aridität und Offenland zu. Bei fortbestehender Dominanz von Steppenverhältnissen zeichnen sich im „Epipaläolithikum“ kürzere klima-ökologische Oszillationen ab. Microtus gud wurde offenbar gebietsweise durch den weniger hygrophilen M. nivalıs als spätem Immigranten von Westen her ersetzt. Mus abbotti erscheint gegen Ende der pleistozänen, Meriones tristrami erst mit holozänen Schichtenfolgen. Einige Arten verschwinden im Jungpleistozän für immer aus dem Gebiet; trotzdem ist eine bedeutende ökologische und faunistische Stabilität zu erkennen. Literatur ALBRECHT, G. (1986): Die Höhlenstation Karain bei Antalya/Türkei. Sondage durch das Institut für Urgeschichte der Universität Tübingen ım Herbst 1985. Mittbl. Archaeol. Venatoria 10/11, 22-36. BESENECKER, H.; SPITZENBERGER, F.; STORCH, G. (1972): Eine holozäne Kleinsäuger-Fauna von der Insel Chios, Agäis (Mammalıa: Insectivora, Rodentia). Senckenbergiana biol. 53, 145-177. CoRBET, G. B.; Morrıs, P. A. (1967): A collection of recent and subfossil mammals from southern Turkey (Asıa Minor), including the dormouse Myomımus personatus. J. nat. Hist. 4, 561-569. FELTEN, H.; SPITZENBERGER, F.; STORCH, G. (1971): Zur Kleinsäugerfauna West-Anatoliens. Teil I. Senckenbergiana biol. 52, 393—424. KRATOCHVIL, J. (1986): Mus abbotti — eine kleinasiatisch-balkanısche Art (Muridae -— Mammalia). Folia Zool. 35, 3-20. KUMERLOEVE, H. (1975): Die Säugetiere (Mammalia) der Türkei. Die Säugetiere (Mammalıa) Syriens und des Libanons. Veröff. zool. Staatssamml. München 18, 69-225. Kuss, $. E.; STORCH, G. (1978): Eine Säugetierfauna (Mammalia: Artiodactyla, Rodentia) des älteren Pleistozäns von der Insel Kalymnos (Dodekanes, Griechenland). N. Jb. Geol. Paläont. Mh. 1978, 206-227. Orsını, PH.; BONHOMME, F.; BRITTON-DAVIDIAN, J.; CROSET, H.; GERASIMOV, $.; 'THALER, L. (1983): Le complexe d’especes du genre Mus en Europe Centrale et Orıientale. II. Criteres d’identification, r&partition et caracteristiques Ecologiques. Z. Säugetierkunde 48, 86-95. OsBorn, D. J. (1962): Rodents of the subfamily Microtinae from Turkey. J. Mammalogy 43, 515-529. — (1965): Rodents of the subfamilies Murinae, Gerbillinae, and Cricetinae from Turkey. J. Egyptian publ. Health Assoc. 40, 401424. RÖTTGER, U. (1986): Schmelzbandbreiten an Molaren der Gattung Arvicola Lacepede, 1799. Unveröffentl. Inaugural-Diss. Med. Fakultät der Rheinischen-Friedrich-Wilhelms-Universität Bonn, 1-121. SPITZENBERGER, F. (1971): Zur Systematik und Tiergeographie von Microtus (Chionomys) nivalıs und Microtus (Chionomys) gud (Microtinae, Mamm.) in S-Anatolien. Z. Säugetierkunde 36, 370-380. — (1972): Der Hamster Mesocricetus brandtı (Nehring, 1898) in Zentralanatolien. Z. Säugetierkunde 37, 229-231. STEINER, H. M. (1972): Systematik und Ökologie von Wühlmäusen (Microtinae, Mammalia) der vorderasiatischen Gebirge Ostpontus, Talysch und Elburs. Sitzungsber. österr. Akad. Wiss., math.-naturwiss. Kl., Abt. I 180, 99-193. £ STORCH, G. (1975): Eine mittelpleistozäne Nager-Fauna von der Insel Chios, Agäıs (Mammalıa: Rodentia). Senckenbergiana bıol. 56, 165-189. — (1977): Die Ausbreitung der Felsenmaus (Apodemus mystacinus): Zur Problematik der Insel- besiedlung und Tiergeographie in der Agäis. Natur u. Museum 107, 174-182. — (1978): Familie Gliridae. In: Handbuch der Säugetiere Europas. Hrsg. J. NIETHAMMER; F. Krapp. Wiesbaden: Akademische Verlagsges. Bd. 1/1, 201-280. Anschrift des Verfassers: Dr. GERHARD STORCH, Natur-Museum und Forschungsinstitut Sencken- berg, Senckenberganlage 25, D-6000 Frankfurt am Main Z. Säugetierkunde 53 (1988) 83-91 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Vocalizations of captive Water mongooses, Atılax paludinosus By CARoLYN M. BAKER Department of Biology, University of Natal, Durban, R.S.A. Receipt of Ms. 11. 2. 1987 Abstract The vocalisations of water mongooses were recorded and analysed sonagraphically. Three main types of sounds were recognised, namely “bray types”, “grizzle types” and “humph types”. “Bray types” were produced during mating and also in some agonistic encounters. “Grizzle types” were produced by both old and young anımals and indicated either distress or a warning. “Humph types” were the only sounds that were characterised by some frequency modulation and were interpreted as either attention or appeasement calls. Compound calls were also recorded and consisted of two or more of the above types produced in rapid succession. It is suggested that the few sounds made by Atilax represent the basıc pattern from which the more complex vocal patterns of sociable herpestines have developed. This suggestion does not, however, assume that the calls of Atzlax are primitive, as they are well adapted to the mongooses way of life. Introduction The variety of vocalisations produced by water mongooses is limited, unlike those of the Indian mongoose Herpestes auropunctatus (MuLLıGan and NerLis 1975), the dwarf mongoose, Helogale undulata rufula (Ma1eEr et al. 1983) the banded mongoose, Mungos mungo (GARRATT 1978) and the suricate Suricata suricatta (EwER 1973). Water mongooses are solitary and nocturnal or crepuscular herpestines. Although relatively silent on their own, the variety of calls produced during social interactions ındicates their ability to communicate adequately when necessary. However, the limited vocal repertoire of Atılax, when compared with the extensive and complex repertoire of sociable herpestines (EwER 1973; GARRATT 1978), indicates that sociable mammals develop a wider vocabulary in response to the demands of group living. In addition to vocalizations, water mongooses communicate behaviourally through facıal expression and body posture as well as through chemical means. Both behavioural and chemical communication will be presented else- where. Material and methods The data presented here were obtained from eight captive water mongooses which were housed either singly or ın pairs in outdoor enclosures measuring 1,5 x 3 x 1,2 m. Details regarding their maintenance are given in BAKER (1987) and BAKER and MEESTER (1986). Recordings were made on an Uher 4000 Report-L tape recorder at a speed of 9,5 cm using either an Uher M517 microphone or a D.K.CG. model D58E directional microphone. Acoustic analysis was performed using a Kay Sonagraph 7029A. For most of the recordings a frequency range of 80 to 8000 Hz was selected, although 40 to 4000 Hz was used for sounds of low frequency. In all cases the narrow filter bandwidth of 45 Hz was selected. Although attempts were made to eliminate background noise during recording sessions, many sonagrams show low frequency noise. The parameters that were described for each sonagram are defined as follows (EISENBERG et al. 1973; RossinG 1982): 1. Fundamental frequency: The component of a sound of the lowest frequency; 2. Harmonics: The components of a sound whose frequencies are multiples of the frequency of the fundamental frequency; 3. Duration: Time of each call or its components (syllables); U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5302-0083 $ 02.50/0 84 Carolyn M. Baker 4. Syllable: Basic call element that manifests itself as an uninterrupted tracing on the horizontal axis of the sonagram; 5. Phrase: Group of syllables separated from other syllables by a time interval greater than any time interval separating the syllables in a phrase; . Tonal syllable: Harmonic syllable; . Noisy syllable: Sound not organised into discrete energy bands; . Mixed syllable: Appears on a sonagram as a superimposition of noise upon a harmonic series; . Long syllable: More than 0,6 seconds in duration; Short syllable: Exceeds 0,05 seconds and less than 0,6 seconds in duration; Formant: A range of frequency to which a system responds preferentially or which is emphasised in ıts output. The context ın which the sounds were made was recorded and briefly described. MOovnuvmoN ON ° . 1 1 Results One hundred and forty-three sounds were sonagraphed and classified into three main types (Fig. 1). Each type was further divided into different forms showing basic similarities. Type one, called “bray types”, consists of long or short mixed syllables which » eo". Noise Energy Bands la KHz TYPE ONE Se el er en 8 2a KHz seconds ——————— 2, TYPE TWO 8 2d KHz seconds ———— 5 8 3a 8 3c 6 6 KHz Al A TYPE THREE 2 2 DCKEEKEREKIKEEHEIEEEEEEN vsoo0000E0 0000000000000 00_ vsoo0000000000 00000000000 » seconds Fig. 1. Three main types of calls produced by Atilax paludinosus Vocalizations of captive Water mongoose, Atılax paludinosus 85 Table 1. Variation in Type 1 calls Type la Type 1b Type lc ANOVA x SD X SD x SD F P Duration 00652926 OO 0.86 0.86 0.79 0.48 Fundamental 1 — 0,98. 0,26 0.8 — 9.63 0.007 frequency Harmonic number 1 - 1 = 14 _ - _ Maximum 020021279 3 A 5.3 oa RD INS frequency exhibit pronounced bands of energy. Type two, called “grizzle types”, are characterised by the presence of pulses. The calls vary in amplitude, and may be long or short and noisy or mixed. Type three, called “humph types” are short, with tonal or mixed syllables. Some frequency modulation is usually apparent. Type one: The three different forms of this call (n = 12) are produced mainly in the context of mating (75 %) and also in some agonistic encounters (25 %). During a mating sequence it is always the female that produces the sound during the initial chase and avoidance phase when she attempts to escape the male. In aggressive encounters it ıs always the subordinate that calls. The posture of this anımal is characteristically submissive with the shoulders at a lower height than the haunches, and often the head is tilted upwards or backwards, pointing towards the dominant anımal. The mouth is usually wide open. Forms la and 1b (Fig. 2) are very sımilar (Table 1) but are significantly different with 84 8- 0.2 0.4 0.6 0.8 1,0 0.2 0.4 0.6 0.8 Seconds —_ nn Seconds Fıg. 2. Type one calls Seconds Signif. S NS S S ANOVA D E 9.62 0.000001 22.09 (02 6.52 0.0008 0.69 0.62 0.73 2.20 3.88 0897 4.58 7.66 0.87 0.64 1.14 3.08 1.47 0.93 0.14 137 2,0%) 0.66 0.31 1.60 0.37 Table 2. Variations in Type 2 calls 0.5 0.31 5 1.81 4.25 0.6 0.97 0.67 7.6 0.85 4.5 1879 9105 0.46 1.04 0.56 0.98 4.54 1.61 Duration Maxımum freq. Formant No. of pulses Carolyn M. Baker respect to their fundamental frequencies (t= 3,92; p= 0,004; df = 8) and maximum frequency (t = 2,51; p = 0,036; df = 8). Form Ic (Fig. 2) is clearly different from 1a and 1b due to the presence of more than one harmonic. However, there are marked sımilarities in duration, maximum fre- quency and fundamental frequency (Table 1). Form 1c was produced exclusively prior to mating. All three of these forms may be repeated within 0,6 seconds, usually depending on the intensity of interaction between two mongooses. Type two: Table 2 shows the main parameters of these forms (n= 94). There are clearly two groups, one of a higher frequency (2a, 2b, 2c) than the other (2d, 2e, 2f) (Fig. 3). The sounds are made equally by both adults and juveniles, although 2c ıs exclusively an adult sound. In most instances (52,1 %) these calls indicated distress, resulting from frustration of goal-directed behaviour. In social encounters they are most commonly produced by a submissive anımal during agonıstic displays (40,4 % of the time). Grizzles are the sounds most commonly found in captive handraised anımals, probably resulting from their restriction. Wild, captive mongooses produce these “griz- zle” sounds (2a, 2b, 2f and 2e) (Fig. 3) only in the context of agonıstic encounters or ın parent-young Interactions. In agonistic encounters they often precede a confrontation and are produced by the submissive anımal. In parent-young interactions the sound may be made by the mother when she is avoiding her offspring, in which case it serves as a warning, or when she is engaged in play-fighting. During these sequences it is the young that call as they are clearly subordinate to the mother. The young “grizzle” when left alone by the mother indicating that they are disgruntled; when playing roughly with each other or their mother as a form of protest; and also when unwilling to be groomed by the parent. In the last two situations the call signals a need to be released. Form 2e (Fig. 3) ıs primarıly produced by very young animals (96,4 % of the time, n=28) and never exhibits pulses. It is a low frequency, noisy sound which may have a formant at 0,89 Hz on average (n= 28, SD = 0,3). It ıs mostly used when the parent leaves the nest for foragıing or when the young have managed to emerge from the nest and are clearly disorientated. Form 2d (Fig. 3) is a growl which is distinct from the others in always being of low frequency, and varyıing widely in duration (Table 2). The pulse frequency is remarkably constant at 0,025 seconds. Of the 14 incidences of growling 14,2 % were made during mating sequences, 50 % were feeding growls and the remainder (35,8 %) were threats made by adults towards the observer. Form 2c (Fig. 3) ıs a cackle produced by adults. It ıs a Vocalizations of captive Water mongoose, Atılax paludinosus 87 8 0.2 0.4 0.6 0.8 1.0 Seconds Seconds 0.2 0.4 0,6 0.8 1.0 1.2 1.3 1.0 Seconds Seconds 307 0.2 0.4 0.6 0.8 1.0 04 08 12 16 20 2.4 Seconds Seconds Fıg. 3. Type two calls loud explosive, clear and concise sound. The growl and the cackle are produced as a threat or when surprised. The pulse interval in type two calls varıies from 0,025 seconds to 0,28 seconds, x = 0,07 seconds, SD = 0,04. Type three: The main parameters of type three calls n = 37 are presented in Table 3. Both young and older animals produce these calls, although 3c is exclusively produced by very young mongooses when callıng for attention. Forms 3a and 3b (Fig. 4) were also made to attract the attention of the parents. In older anımals 3a and 3b were often produced in anticipation of food. During mating sequences the male was heard callıng 88 Carolyn M. Baker Table 3. Variations in Type 3 calls Type 3a ANOVA x SD D SD B Signif. Duration 0.28 0.04 ß 0.37 NS Maximum 4.58 1.49 Ä O.OOOEEEES frequency Fundamental frequency Harmonic No. Max. freq. modulatıon KHz {| Seconds Seconds Fıg. 4. Type three calls 0.4 Seconds when confronted with an unwilling female. Because this sound was made only in response to a female threatening a persistent mate (in the context of mating) during the initial phases of the mating sequence, it indicates an appeasement function. Form 3c exhibits a marked frequency modulation (Fig. 4; Table 3), and is the only water mongoose call to do so. Occasıonally different sounds were grouped together to form a compound call. Most commonly “grizzle” sounds were repeated to form a phrase, although some sounds consisted of a series of the above types following one another sequentially, or superim- posed one upon the other (Fig. 5). Most commonly there was a combination of types two and three, beginning with a “grizzle” and ending with a “humph”. Type two (“grizzle”) might also be followed by a type one sound (“bray”). Vocalizations of captive Water mongoose, Atılax paludinosus 89 A rs ir nr ” KM ee De z KL en % " % 0.4 0.4 Seconds Seconds Fig. 5. Compound calls produced by Atılax. A: Grizzle followed by humph; B: Grizzle followed by bray Discussion Although the variety of vocal sounds made by captıve Atzlax ıs poor, their inherent variability is extensive, especially when duration and maximum frequency are considered. The only other major variant is the number of pulses that occurs in Type 2 calls. This is clearly related to duration, as the longer the calls the greater the number of pulses. The variatıon in pulse intervals shows no clear pattern, and is neither sex- nor age-related, as both young and old anımals exhibit both extremes of the range. EISENBERG et al. (1973) discuss varıability within the categories of sounds produced by marsupials, and note that a number of transıtional forms of the defined type may exist, such that the distinctness of two types becomes confused. Further, the sounds that are used as the categorical type are usually at the extremes of a graded series of sounds, and thus are quite distinct. In Atzlax there are some sounds that may structurally appear to fit into one category or another, such as types la, 2a and 3a sounds. It ıs often only the auditory perception of the sound in conjunction with its visual context that provides the key to the category ın which it belongs. The variation in Atzlax calls may have one or more of several determinants, most notably individual varıation or age-related differences. In addition, a solitary mongoose with its limited vocal repertoire demands less specificity with respect to the sounds that it makes, as there is little chance of confusion when so few calls make up the entire vocabulary. In a social anımal with a wider range ot different calls it becomes important to ensure correct understanding of specific signals, as these are crucıal in maintaining group understanding. Therefore it would be expected that greater attention would be paid to pronouncing the sounds concisely. The tendency ın colonıal birds (WıLEY 1976) ıs towards reduction ın the varıety and detail of calls as these become confused in the noisy environment of a breeding colony. This may be an important factor influencing the vocalısations of mammals, and would be most clearly illustrated amongst some of the sociable herpestines. However, the available information on vocal repertoires of sociable mongooses indicates a deviation from this pattern, which may also be representative of other mammalian groups. When compared with vocalisations of Mungos mungo (GARRATT 1978) and Herpestes auropunctatus (MULLIGAN and NELLIS 1975) it becomes clear that the sounds made by Atılax are a basıc type that ıs found within the repertoire of these other two species, and which has in some instances been slightly modified (e.g. the weeonk call of Herpestes 90 Carolyn M. Baker auropunctatus resembles the “humph” call of Atzlax) or elaborated upon (e.g. the squeal and “we” calls of Mungos mungo share similarıties with the “humph” of Atzlax). By virtue of their solitary nature water mongooses have no need of a “moving out call” or a contact call (MAIER et al. 1983) as these are particularly related to aspects of group life. The sounds produced by the water mongoose may represent those that are essential for maintaining compatibility during their comparatively brief encounters and perhaps it is not surprising that the sounds made most commonly by adults were related to either agonistic behaviour or distress. The captive conditions are likely to have induced a greater usage of these two kinds of sounds, and in the natural environment their incidence may be greatly reduced. It was only the “humph” call that elicited an amicable response in captivity, although some of the type 2 calls (Type 2e) made by juveniles were used to elicit a positive response from either siblings or parent. In general the sounds made by Atılax are of low frequency and are remarkably simple, and it ıs suggested that they represent an example of the basic pattern from which the more complex vocal repertoires of other herpestines have developed. BusneL (1963) supports this suggestion, and comments that the number of signals made by anımals vary, but that in general their number increases ‘the higher one goes in the Anımal Kingdom’. Nevertheless the fact that the sounds made by Atzlax may represent a basic pattern does not necessarily mean that they are primitive, as water mongooses are clearly well adapted to their environment. Indeed, Atılax has developed specialised, but simple techniques for optimis- ing survival in its particular niche. Acknowledgements Thanks are due to the following people: Professor J. MEESTER for supervising the project and reading the manuscript; Ms A. NoLTtE and Dr P. WARERLEY for allowing access to the sonagraphic equipment in the Department of Physics; Professor ©. A. E. Rasa and Mr D. Basckın for reading and commenting on the manuscript; Professor G. MACLEAN for translating the summary; Mr D.M. Drammmnt for preparing photographs of the sonagrams. Financial assistance from the South African Council for Scientific and Industrial Research and the University of Natal, Durban, is gratefully acknowledged. Zusammenfassung Vokalisationen des Sumpfichneumon, Atılax paludinosus, in Gefangenschaft Die Lautäußerungen von Sumpfichneumons in Gefangenschaft wurden aufgenommen und sonagra- phisch analysiert. Drei Haupttypen von Lauten konnten unterschieden werden: Schreien, Murren und „Hm“-Laute. Schreilaute wurden hauptsächlich während der Paarung geäußert, aber auch in einigen agonistischen Verhaltensweisen. Laute des Typs Murren wurden von alten und auch jungen Tieren wiedergegeben, entweder in einer Notlage oder als Warnung. „Hm“-Laute, die einzigen Laute, die durch Frequenz-Modulation charakterisiert sind, dürfen als Achtungs- oder Beruhigungs- laute bezeichnet werden. Zusammengesetzte Rufe wurden ebenfalls analysiert. Sie bestanden aus zwei oder mehr der oberen Typen, die schnell nacheinander geäußert wurden. Es wird angenommen, daß die wenigen Lauttypen des solitären Atzlax ein Basis-Muster darstellen, aus dem sich die komplizierte- ren Muster sozialer Herpestinae entwickelt haben könnten. Diese Annahme sollte jedoch nicht zu der Aussage verleiten, daß die Atzilax-Rufe primitiv seien, denn sie sind der Lebensweise dieser Mangu- stenart gut angepaßt. Literature BAKER, C. M. (1987): The biology of the water mongoose, Atılax paludinosus. PhD. Thesis (in prep.). BAKER, C. M.; MEESTER, J. (1986): Postnatal physical development of the water mongoose (Atılax palndinosus). 7 Säugetierkunde 51, 236-242. BusnEL, R. G. (1963): Aspects of nmel nasse signals. In: Acoustic behaviour of anımals. Ed. by R. G. Busner. London: Elsevier. EISENBERG, ]. F.; Coins, L. R.; WEMMER, C. (1973): Communication in the Tasmanıan Devil (Sarcophilus harrisii) and a survey of auditory communication in the Marsupialia. Z. Tierpsychol. 37, 379-399. EwER, R. F. (1973): The carnıvores. London: Weidenfeld and Nicolson. Vocalizations of captive Water mongoose, Atılax paludinosus 91 GARRATT, P. A. (1978): Vocalızations of the banded mongoose (Mungos mungo). Honours Thesis, Durban: Univ. Natal. MAIER, V.; Rasa, ©. A. E.; SCHEICH, H. (1983): Call-system sımilarıty in a ground-living social bird and a mammal in the bush habitat. Behav. Ecol. Sociobiol. 12, 5-9. MuLLIGan, B. E.; Nerııs, D. W. (1975): Vocal repertoire of the mongoose Herpestes auropunctatus. Behaviour 55, 237-267. Rossıng, T. D. (1982). The science of sound. Reading: Addison-Wesley. Wırey, R. H. (1976): Communication and spatial relationships in a colony of common grackles. Anım. Behav. 24, 570-584. Author’s address: CaRroOLYN M. BAKER, Department of Biology, University of Natal, King George V Avenue, Durban 4001, R.S.A. Z. Säugetierkunde 53 (1988) 92-101 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Secondary sex ratio in the Przewalski horse Equus przewalskii (Mammalia: Equidae) By J. MLikovskf Department of Evolutionary Biology of the Czechoslovak Academy of Sciences Receipt of Ms. 16. 3. 1987 Abstract Sex ratio distribution in sibships and the relation between the secondary sex ratio and maternal age, paternal age, parity, birth sequence, season and year were studied in the Przewalski horse Equus przewalskı. No varıations of the secondary sex ratio in relation to the studied factors were detected; sex ratio distribution was found to be simply binomial, i.e. random. The overall secondary sex ratio was found to be not significantly different from unity both in Przewalski horses and in other perissodactyls. These data neither support, nor directly contradict, the theory that females are capable of adjusting the sex ratio of their offspring. Introduction Variations in the secondary sex ratio represent an important component of the TRIvERS and WILLARD’s (1973) hypothesis of the adaptıve value of the female’s ability to adjust facultatively the sex of its offspring (see also BuURLEY 1982), which in turn forms a part of the parental investment theory (TRIvERS 1972; MAYNARD SMITH 1977, 1980; CLUTTON- Brock and Arson 1982), and are thus of considerable theoretical interest. Despite this, data relevant to the problem are still rather scarce, even in mammals (CLUTTON-BROcK and Aıson 1982; BoL’Ssakov and Kusancev 1984; CLUTTON-BRocK 1986a) and birds (CLUTTON-Brock 1986b). In the present paper I will contribute to the problem under discussion by analyzing the secondary sex ratio in the Przewalski horse Eguus przewalskü Poljakov, 1881. Captive Przewalski horses seem to be suitable for this purpose for the following reasons: They are strictly monotocous (MoHR and VOLF 1984), so that varıations in the litter sıze are ruled out (cf. SMITH and FRETwELL 1974; Myers 1978; GosLing 1986b); they are free of nutritional stresses which were found to be important in this respect (cf. RıvERS and CRAWFORD 1974; MCCLurRE 1981; VAN DER MERWE and SKINNER 1982; SıLk 1983; VERME 1983, 1985; McGinLEy 1984); and they do not suffer from predation (cf. BERGER 1983). It could be thus expected that the effect of the remaining factors on the secondary sex ratio will be more pronounced and easier to detect. Material and methods All of the demographic data used in the present paper were extracted from the published Pedigree Books of the Przewalski Horse (VoLr 1980, 1981, 1982, 1983, 1984). They are, hence, considering captive anımals. Overall, data on 1115 newborn foals were obtained, produced by 262 mares and 113 staillons. In testing the results, standard statistical procedures were employed (SokAL and ROHLF 1969; Sachs 1974). U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5302-0092 $ 02.50/0 Secondary sex ratio in Equus przewalskii 93 Results Overall secondary sex ratio Of the 1115 newborn Przewalski foals of known sex registered in the Pedigree Books 529 (= 47.44 %) were males and 586 (= 52.56 %) females. The slight preponderance of females is statistically not significant (see Tab. 1). To see whether this is a specific feature of Przewalski horses, or whether the 1:1 secondary sex ratio is typical for horses and other perissodactyls, I compiled the relevant data from the “Mammals bred in captivity and multiple generation births” section of the International Zoo Yearbook, vols. 1-23 (Tab. 1). The data show convincingly that ın not one of the listed species the secondary sex ratio significantly deviates from unity, although rhinos posses some tendency to bear more males than females (p = 0.065). Effect of maternal age The ability to parental investment varies with the female’s age (CaswELL 1982, 1984), so that it could be expected that the secondary sex ratio will vary with the female’s age as well. Indeed, such a variation has been reported for European bisons Bison bonasus (ZABLOCKIJ 1957), red deer Cervus elaphus (LowE 1969), white-tailed deer Odocozleus virginianus (VERME 1983, 1985), barbary macaques Macaca sylvanus (PauL and THOMMEN 1984), Table 1. Overall secondary sex ratio in wild (w), feral (f) and captive (c) Perissodactyla Taxon N INIE (96.6) NPZ2002) Nd CH; References Equus przewalskiü 1115 529(47.44) 586 (52.56) 1.358 this study Equus caballus 476 248 (52.10) 228 (47.90) S 0.341 WELSH 1975; FEısT and MCcCULLOUGH 1975, 1976; BER- GER 1977, 1983; KEıpER 1979; NELsSon 1980 400 (51.82) 372(48.19) 1. 438 this study 36 (50.00) 36(50.00) 1. .028 this study 160.0) Oo) © .113 McCorr 1979 194130) 27 (58.70) h 5 PENZHORN 1975 141 (51.09) 135(48.91) 1. .029 this study 114 (52.78) 102(47.22) 1. .232 Smurs 1976 Equus quagga 1340 (47.98) 1453 (52.02) O0. this study Equus grevyi SD IB) FIN AESODE N: 3 this study Equidae 6411 3157 (49.24) 3254 (50.76) Tapirus indicus 120 61 (50.83) SAID Te : this study Tapirus terrestris 2255045) 229 5I 1. 5 this study Tapirus bairdi 12 (63.16) 7 (36.84) 1. this study Tapirus pinchaque 3 (50.00) IS) this study Tapiridae 301 (50.93) 290 (49.07) Rhinoceros unicornis 24 (64.86) ISIESrA) this study Ceratotherium simum 87 (58.78) 61 (41.22) 1. this study Diceros bicornis 63 (54.31) BI) 1 this study Rhinocerotidae 174 (57.81) 127 (42.19) 1.370 Equus hemionus Equus africanus Equus asınus Equus zebra Equus zebra Equus quagga anidindmnn N = number of newborn animals sexed. G,4; = Woorr’s (1957) log likelihood ratio test, adjusted after YATES, two-tailed. No deviations are significantly different from unity (p > 0.05). 94 J: Mlikovsky captive coypus Myocastor coypus (GosLinG 1986a), captive foxes Alopex lagopus (ZEGALOV 1950), varıous species of domestic mammals (see BoL’$sakov and KuBANcEV 1984 for review) and ın man (e.g., MLikovskY 1987). On the other hand, no such variation was revealed in Przewalski horses (this study, Tab. 2) and in rhesus monkeys Macaca mulatta (Rawrins and KessLEr 1986). Effect of paternal age In equids, males do not care for foals (GrovEs 1974), so that no varıation in the secondary sex ratio with the paternal age ıs to be expected on the basis of the parental investment theory. However, other mechanisms, such as physiological, may cause this relationship. The results show that the secondary sex ratio does not vary with the paternal age in Przewalski horses (Tab. 3), which agree in this respect with red deer (CLUTTON-BROcK et al. 1981) and domestic mice Mus musculus (SUNTZEFF et al. 1962). However, such a varıatıon was described in pıne martens Martes martes (GRAKOV 1969), various species of domestic mammals (KamALjJan 1962; BoL’Sakov and KuBAncEv 1984) and in man (e.g., PoLLArD 1969; MLikovskY 1987). Effect of parity The effect of parity on the secondary sex ratio has been little studied thus far and seems to have no direct theoretical interpretation. Przewalskı horses showed no variation of the secondary sex ratio with parity (Tab. 4). The same result has been reported also for rhesus monkeys (Rawıins and KesstLer 1986) and domestic pigs Sus scrofa (KENNEDY and Moxıry 1978; Gray and KATANBArF 1985). Table 2. Relation between the secondary sex ratio and maternal age (at delivery) in Equus przewalskii, based on the offspring of 262 mares NE (% 8) N? (% 2) 1 28 (46.67) 54 (51.92) 60 (51.28) 49 (45.37) 44 (48.35) 37 (41.66) 30 (38.46) 46 (63.01) 29 (42.03) 26 (45.61) 21 (46.67) 23 (46.94) 15 (51.72) 12 (31.58) 12 (54.55) 7 S1.82) 13 (56.52) 13 (48.15) 32 (53.33) 50 (48.08) 57 (48.72) 59 (54.63) 47 (51.65) 54 (59.34) 48 (61.54) 27 (36.99) 40 (57.97) 31 (54.39) 24 (53.33) 26 (53.06) 14 (48.28) 26 (68.42) 10 (45.45) 15 (68.18) 10 (43.48) 14 (51.85) N = number of sexed newborns. x” test, adjusted after YATEs, two-tailed. No deviations from the mean sex ratio are statistically significant (p> 0.05), excepting in 10 year old females, which tended to bear more sons than daughters (0.05>p>0.01). However, the 9 year old females showed the opposite (though statistically insignificant) tendency and after combining the two years together, no more deviation was apparent (764 : 759; za; = 0.397; n.s.). Such seeming statistical deviations are expected to occur on statistical grounds (Hırı 1985). Secondary sex ratio in Equus przewalskü 95 Table 3. Relation between the secondary sex ratio and paternal age (at foal’s birth) in Equus przewalskii, based on the offspring of 113 stallions NE (% 8) N? (% 2) 2 (25.00) 16 (47.06) 38 (48.10) 47 (48.96) 48 (47.52) 6 (75.00) 18 (52.94) 41 (51.90) 49 (51.04) 53 (52.48) 49 (53.85) 50 (46.30) 50 (53.76) 46 (60.53) 41 (62.12) 23 (41.07) 27 (55.10) 21 (52.50) 8 (50.00) 10 (55.56) N = number of sexed newborns. x” test, adjusted after YATEs, two-tailed. No deviations from expected distributions are statistically significant (p > 0.05). Birth Nd (% 8) N? sequence 125 (48.08) 99 (50.77) 81 (51.27) 49 (40.16) 46 (46.00) 37 (45.12) 31 (51.67) 24 (52.17) 15 (45.45) 10 8 (38.10) 15 14 (43.75) 18 (56.25) DON PRWMN HT N = number of sexed newborns. x” test, adjusted after YATEs, two-tailed. No deviations from the mean sex ratio are statistically significant (p > 0.05). Effect of birth sequence Sex sequences in ındıvidual families were investigated ın man only so far, where they were found to be random in several studies (RENKONEN 1956; EpwArnDs 1961, 1962; LoyD and GraY 1969) and non-random in other ones (SCHÜTZENBERGER 1949, 1950; GRAY and MoRrRISon 1974). No deviation from randomness was revealed in Przewalski horses, where sex sequences were studied ın famılies of 122 females with 4-15 foals (e = 0.184, p >0.1; non-parametric run test after WaLp and WorLrowttz 1940; cf. Krar 1983). 96 J: Mlikovsky Effect of the female’s predisposition The female’s predisposition to bear either more daughters or more sons can be detected by an analysıs of the sex ratio distribution in litters or families (sibships). The expected statistical distribution of the sex ratio within litters or families is the binomial one (Epwarps 1960, 1962; HARMSEN and CookeE 1983). The randomness of the sex ratio distribution in lıtters or families was corroborated in Przewalski horses (this study, Table 5. Sex ratio distribution in sibships of Equus przewalskii [EN +0 Sıbship No. of No. of litters size litters observed expected 2 43.89 37.23 53.87 18.27 58415 ON PERUMNDHMTO NN NEON HO OS T RPODHO T$$UNDH-O U NDNHO DDHO N, 60o 2: 1: 02 3): 722 108 0: 4: 3% D3: 1% (0)5 538 4: 3% 2: 118 0% 6: sr 4: 3): 24: 1: 08 7£2 6: 5: 4: 98 Da: 1: 0: 8: 7a: 6: 5E 4: 35 2>* ie 0: X test, two-tailed. No deviations from expected distributions are statistically significant (pP > 0.05). Secondary sex ratio in Equus przewalskıu 97 Tab. 5), dairy cattle Bos taurus (Gray and Hurt 1979), Siberian tigers Panthera tıgris altaica (MıfkovskY 1985) and in several studies on man (EDwArDs and FRAccARoO 1958, 1960; Gray and Morrıson 1974; Gray and BorToLozz1 1977). On the other hand, some controversial indications exist that the sex ratio distribution in litters or families deviates from the binomial one in man (GEISSLER 1889; EDwArDs 1958), coypus (GosLInG 1986a), domestic pigs (Gray and KATAnBAF 1985) and in several other species of domestic mammals (Ginı 1951; James 1975). Effect of the female’s body condition Direct data on body condition of pregnant Przewalski mares were not available, but TRIvErs and WILLARD (1973) have assumed that in large mono- or oligotocous mammals, the female’s body condition is higher when the female has skipped reproduction in the foregoing year. Using this measure, no effect of the female’s body condition on the secondary sex ratio of that female’s offspring was detected in Przewalski horses (Tab. 6) but, using another measure, SKOGLAND (1986) revealed this effect in reindeer Rangıfer tarandus. Table 6. Relation between the secondary sex ratio and reproductive effort (measured as number of foals born to 3-6 year old mares) in Equus przewalskii N? (% 9) 7 (45.00) E (53.78) “ (52.38) ) 9 (56.25 Mares which died before 6 year old were excluded from the analysıs. N = number of sexed newborns. x? test, adjusted after YATEs, two-tailed. No deviations from the mean sex ratio are statistically significant (p > 0.05). Seasonal varıations There is a distinct, though extended, foalıng season in captive Przewalski horses ın the Palearctic (MoHRr and Vorr 1984), where 1103 of the 1115 registered Przewalski foals were born. Hence, it could by hypothesized that the secondary sex ratio will show some seasonal variation. However, no such varıation was detected (Tab. 7). BAUMGARTNER (1985), who studied the same relationship in domestic hens Gallus gallus f. domestica, has found such a variation in one of the two strains he studied, but not in the other one. On the other hand, a strong seasonal varıation in the secondary sex ratio was found in grey seals Halichoerus grypus (CouLson and HickLıng 1961) and in captive coypus (GOSLING 1986a). Annual variations Year to year variations in the secondary sex ratio were reported for a number of mammalian species, including, for example, American buffalos Bison bison (McHucH 1959), European bisons (KOROCKINA 1968) and water buffalos Bubalus bubalıs (FISCHER 1966; Coswamı and NAıR 1968). However, no such varıation was observed in Przewalski horses (this study, Tab. 8). 98 J. Mlikovsky Table 7. Relation between the secondary sex ratio and season in Equus przewalskii Month I II III IV V VI VII VII IX x x] XI N = number of sexed newborns. x” test, adjusted after YATEs, two-tailed. No deviations from the mean sex ratio are statistically significant (p > 0.05). Period 1901-10 1911-20 1921-30 193140 1941-50 1951-60 1961-65 1966-70 1971-75 1976-80 1981-83 Nd (% 8) 9 (52.94) 6 (46.15) NS (% 8) 5 (38.46) 13 (40.63) 15 (41.67) 23 (60.53) 10 (40.00) 31 (44.93) 45 (50.00) 58 (55.77) 71 (43.83) 138 (46.46) 124 (50.41) N? (% 2) 8 (47.06) 7 (46.67) 15 (42.86) 81 (54.73) 182 (50.84) 141 (54.23) 67 (54.47) 27 (49.09) 21 (51.22) 12 (44.44) 6 (54.55) 7 (53.85) N? (% 2) 8 (61.54) 19 (59.37) 21 (58.33) 15 (39.47) 15 (60.00) 38 (55.07) 45 (50.00) 46 (44.23) 91 (56.17) 159 (53.54) 122 (49.59) N = number of sexed newborns. x? test, adjusted after YATEs, two-tailed. No deviations from the mean sex ratio are statistically significant (p > 0.05). Discussion Whether adaptive adjustment of the secondary sex ratio occurs in birds and mammals is still inconclusive (WırLıams 1979; CHARNOV 1982; CLUTTON-BRocK and ALBoN 1982), because the relevant data are highly controversial. Since many factors may directly or indirectly affect the secondary sex ratio in higher vertebrates, ıt is favorable to study its variations experimentally (e.g., MCCLure 1981; Myers et al. 1985; LaBov et al. 1986) and/ or to select for the study such species or populations which are naturally independent of some of the possible factors (e.g., RUTBERG 1986). As shown above, captıve Przewalski horses are suitable for such a kind of research. Despite that, none of the investigated factors, viz. maternal age, paternal age, parity, birth sequence, female’s predisposition, female’s body condition, season and year, proved effective. To conclude, nothing was found in the varıations of the secondary sex ratio in the Przewalski horse in support of the Trıvers and Wırrarn’s (1973) sex adjustment hypothesis. However, it should be noted at the same time that two other important factors were not investigated in the present study, particularly the effect of the time of fertilization Secondary sex ratio in Equus przewalskü 99 within the estrus cycle (James 1976, 1983; VERME and OzocA 1981) and the effect of the social status of the female in the herd (Sırk et al. 1981; Stmpson and Sımpson 1982; CLUTTON-BRocK 1982; CLUTTON-BrRocK et al. 1984, 1986). Hence, the results reached, though not supporting the TRıvErs and WILLARD’s (1973) hypothesis, cannot be used as a proof of its incorrectness. Zusammenfassung Das sekundäre Geschlechterverhältnis beim Przewalski-Pferd, Equus przewalskii (Mammalia: Equidae) Das sekundäre Geschlechterverhältnis, seine Verteilung in Familien und die Beziehungen zwischen dem sekundären Geschlechterverhältnis und dem Alter von Mutter und Vater, der Parıtät, Geburts- folge, Jahreszeit und Jahr wurden bei Przewalski-Pferden untersucht. Die Verteilung des Geschlech- terverhältnisses wich von einer Binomialverteilung nicht signifikant ab. In dem sekundären Geschlechterverhältnis wurden keine Variationen entdeckt. Es wich weder bei Przewalski-Pferden noch bei anderen Perissodactylen signifikant von 1:1 ab. Diese Angaben unterstützen weder die Theorie, daß die Weibchen das Geschlecht ihrer Nachkom- men beeinflussen können, noch widersprechen sie ihr direkt. Literature BAUMGARTNER, J. (1985): Vplyv genotypu, sezönnosti a doby liahnutia na sekundärny pomer pohlavia kury domäce) (The influence of genotype, season and hatching time on the secondary sex ratio of domestic fowl). Zivo£is. Vyroba 30, 1083-1092. BERGER, J. (1977): Organızational systems and dominance in feral horses in Grand Canyon. Behav. Ecol. Sociobiol. 2, 131-146. — (1983): Predation, sex ratios, and male competition in equids (Mammalia: Perissodactyla). ]. Zool., London, 201, 205-216. Bor’Sarov, V. N.; KuBancev, B. S. (1984): Polovaja struktura populjacij mlekopitajuscich ı e& dinamika (Sex structure of mammalıan populations and its dynamics). Moskva: Nauka. BurLe£y, N. (1982): Facultative sex-ratio manıpulation. Amer. Nat. 120, 81-107. CAswELL, H. (1982): Optimal life histories and the age-specific costs of reproduction. J. theor. Biol. 98, 319-329. — (1984): Optimal life histories and age-specific costs of reproduction: two extensions. J. theor. Biol. 107, 169-172. CHARNOoVv, E.L. (1982): The theory of sex allocation. Princeton: Princeton Univ. Press. CLUTTON-Brock, T. H. (1982): Sons and daughters. Nature 298, 11-13. — (1986a): Sex ratio variatıion in mammals. Quart. Rev. Biol. 61, 339-374. — (1986b): Sex ratio varıatıon in birds. Ibis 128, 317-330. CLUTTON-BrRock, T. H.; Arson, $. D. (1982): Parental investment in male and female offspring in mammals. In: Current problems in sociobiology. Ed. by King’s Coll. Sociobiology Group. Cambridge: Cambridge Univ. Press. pp. 223-247. CLUTTON-BRocK, T. H.; Aıson, $. D.; Guinness, F. E. (1981): Parental investment in male and female offspring in polygynous mammals. Nature 289, 487—489. — (1984): Maternal dominance, breeding success and birth sex ratios in red deer. Nature 308, 358-360. — (1986): Great expectations: dominance, breeding success and offspring sex ratios in red deer. Anım. Behav. 34, 460-471. Coswamı, $. B.; Naır, A. P. (1968): A comparative study of secondary sex ratio, twinning and multiple births in the Indian water buffalos and other farm animals. Indıan J. Dairy Sci. 21, 50-56. Couson, J. C.; HıckLıng, G. (1961): Variation in the secondary sex-ratio of the grey seal Halichoerus grypus (Fab.) during the breeding season. Nature 190, 281. Epwarps, A. W. F. (1958): An analysıs of Geissler’s data on the human sex ratio. Ann. Human. Genet. 23, 6-15. — (1960): The meaning of binomial distribution. Nature 186, 1074. — (1961): A factorıal analysıs of sex ratio data. Ann. Human Genet. 25, 117-121. — (1962): A factorıal analysıs of sex ratio data: a correction to the article in vol. 25, 117. Ann. Human Genet. 25, 343-345. EDwarDs, A. W. F.; FRACcARo, M. (1958): The sex ratio distribution in the offspring of 5477 Swedish ministers of religion, 1585-1920. Hereditas 44, 447—450. Feist, J. D.; McCurroucH, D. R. (1975): Reproduction in feral horses. J. Reprod. Fert. (Suppl.) 23, 13-18. — (1976): Behavior patterns and communication in feral horses. Z. Tierpsychol. 41, 337-371. 100 J. Mlikovsky FiscHER, H. (1966): Secondary sex ratio in the water buffalo (Bubalus bubalıs). Z. Tierzüchtg. Züchtgsbiol. 82, 361-363. GEISSLER, A. (1889): Beiträge zur Frage des Geschlechterverhältnisses der Geborenen. Z. kgl. sächs. statıst. Bureau 35, 1-24. Gırr, J. L. (1985): Interpretation of significance in testing multiple traits. J. Anim. Sci. 60, 867-870. Gin, ©. (1951): Combinations and sequences of sexes in human families and mammal litters. Acta genet. statist. med. 2, 220-244. Gosring, L. M. (1986a): Selective abortion of entire litters in the coypu: adaptive control of offspring production in relation to quality and sex. Amer. Nat. 127, 772-795. — (1986b): Biased sex ratios ın stressed anımals. Amer. Nat. 127, 893-896. GRAKoV, N. 1. (1969): Polovaja, vozrastnaja struktura ı prirost populjacii lesnoj kunicy (Sex and age structure and population growth ın the pine marten). In: Voprosy &kologii promyslovych zivotnych (Questions of the ecology of game anımals). Kirov: VNIIZP. pp. 3-18. Gray, E.; BORTOLOZZT, J. (1977): Studies of the human sex ratio and factors influencing family size in Botucatu, Brazil. J. Hered. 68, 241-244. Gray, E.; Hurt, V.K. (1979): Distribution of sexes in cattle. J. Hered. 70, 273-274. Gray, E.; KATAnBAF, M.N. (1985): Sex ratio and distribution of sexes in swine. J. Hered. 76, 36-38. Gray, E.; MoRRISoN, N. M. (1974): Influence of combinations of sexes of children on family size. ]J. Hered. 65, 169-174. Grovzs, C. P. (1974): Horses, asses, and zebras in the wıld. Hollywood: Curtis Books. HarMmsEn, R.; CookE, F. (1983): Binomial sex-ratio distribution in the lesser snow goose: a theoretical enigma. Amer. Nat. 121, 1-8. Jamzs, W. H. (1975): The distributions of the combinations of the sexes in mammalıan litters. Genet. Res. 26, 45-53. — (1976): Timing of fertilization and sex ratio of offspring. A review. Ann. Human Biol. 3, 549-554. — (1983): Timing of fertilization and the sex ratio of offspring. In: Sex selection of children. Ed. by N. G. BEnnETT. Orlando: Academic Press. pp. 73-100. Kamaıjan, V. 5. (1962): Vlijanie vozrasta roditele) na sootnosenie pola potomstva (Effect of the parental age on the sex ratio of offspring). Z. obs£. Biol. 23, 455-459. KEIPER, R. R. (1979): Population dynamics of feral ponies. In: Symposium on the ecology and behavior of wild and feral equids. Ed. by R. H. Denniston. Laramie: Univ. Wyoming. pp. 175-184. Kennepy, B. W.; Moxtevy, J. E. (1978): Genetic and environmental factors influencing litter size, sex ratio and gestation length in the pig. Anım. Product. 27, 35-42. Kıart, M. (1983): Sex of children and family size: a decision model. Math. Bioscı. 65, 171-185. Korockına, L. N. (1968): Faktory, vlijajuscie na sootnoSenie pola u zubrov (Factors influencing sex ratio ın bisons). Belovezskaja pusca (Beloveza forest) 2, 139-147. Lagov, J. B.; Huck, U. W.; Vaswantı, P.; Lısk, R. D. (1986): Sex ratio manıpulation and decreased growth of male offspring of undernourished golden hamsters (Mesocricetus auratus). Behav. Ecol. Sociobiol. 18, 241-249. Lowe, V. P. W. (1969): Population dynamics of the red deer (Cervus elaphus L.) of Rhum. J. anım. Ecol. 38, 425-457. Loyp, R. C.; Gray, E. (1969): A statistical study of the human sex ratio. J. Hered. 60, 329-331. MAYNARD SMITH, ]. (1977): Parental investment: a prospective analysis. Anım. Behav. 25, 1-9. — (1980): A new theory of sexual investment. Behav. Ecol. Sociobiol. 7, 247-251. MCcCLure, P. A. (1981): Sex-biased litter reduction in food-restricted wood rats (Neotoma floridana). Science 211, 1058-1060. McCorrt, W. D. (1979): The feral asses (Equus asinus) of Ossabau Island, Georgia: mating system and the effect of vasectomies as a population control procedure. In: Symposium on the ecology and behavior of wild and feral equids. Ed. by R. H. Dennıston. Laramie: Wyoming. pp. 71-83. McGintey, M. A. (1984): The adaptıve value of male-biased sex ratios among stressed anımals. Amer. Nat. 124, 597-599. McHuchn, T. (1958): Social behavior of the American buffalo (Bison bison). Zoologica 43, 29-32. MLfkovskf, J. (1985): Sex ratio distribution in the Siberian tiger Panthera tigris altaica (Mammalıa: Felidae). Z. Säugetierkunde 50, 47-51. — (1987): On the relationship between the secondary sex ratio and parental age in man. Acta Univ. Carol. (Med.), (in press). MonR, E.; Vorr, J. (1984): Das Urwildpferd Equus przewalskü. Wittenberg Lutherstadt: A. Ziemsen Verlag. Myers, J. H. (1978): Sex ratio adjustment under food stress: maximization of quality or number of offspring? Amer. Nat. 112, 381-388. Myers, P.; MASTER, L. L.; GARRETT, R. A. (1985): Ambient temperature and rainfall: an effect on sex ratio and litter size in deer mice. J. Mammalogy 66, 289-298. Neıson, K. ]J. (1980): Sterilization of dominant males will not limit feral horse populations. U.S.D.A., For. Serv. Res. Pap. RM-226. Secondary sex ratio in Equus przewalsku 101 Pau, A.; THomMmEn, D. (1984): Timing of birth, female reproductive success and infant sex ratio in semi-free-ranging Barbary macaques (Macaca sylvanus). Folia primatol. 42, 2-16. PENZHORN, B. L. (1975): Behaviour and population ecology of the Cape mountain zebra, Equus zebra zebra L., 1758, in the Mountain Zebra National Park. Unpubl. PhD. thesis, Univ. Pretoria. PoLLarp, G. N. (1969): Factors influencing the sex ratio at birth in Australia, 1902-1965. J. biol. Scı. 1, 125-144. Rawrins, R. G.; KessLer, M. J. (1986): Secondary sex ratio variation in the Cayo Santiago macaque population. Amer. J. Primatol. 10, 9-23. RENKONEN, K. ©. (1956): Is the sex ratio between boys and girls correlated to the sex of precedent children? Ann. Med. exp. Fenn. 34, 447451. Rivers, J. P. W.; CRAwFoRD, M. A. (1974): Maternal nutrition and the sex ratio at birth. Nature 252, 237298. RUTBERG, A. T. (1986): Lactation and fetal sex ratios in American bison. Amer. Nat. 127, 89-94. Sachs, L. (1974): Angewandte Statistik. Berlin: Springer-Verlag. SCHÜTZENBERGER, M. P. (1949): Resultats d’une enqu£te sur la distribution du sexe dans les familles nombreuses. Sem. Höp. Parıs 25, 2579-2582. — (1950): Nouvelles recherches sur la distribution du sexe ä la naissance. Sem. Höp. Paris 26, 4458-4465. Sırk, J. B. (1983): Local resource competition and facultative adjustment of sex ratios in relation to competitive abilities. Amer. Nat. 121, 56-66. SILK, J. B.; CLARK-WHEATLEY, C. B.; RODMan, P. $.; SAMUELS, A. (1981): Differential reproductive success and facultative adjustment of sex ratios among captive female bonnet macaques (Macaca radiata). Anım. Behav. 29, 1106-1120. Sımpson, M. ]J. A.; Sımpson, A. E. (1982): Birth sex ratios and social rank in rhesus monkey mothers. Nature 300, 440-441. SKOGLAND, T. (1986): Sex ratio variation in relation to maternal condition and parental investment in wild reindeer Rangıfer t. tarandus. Oikos 46, 417419. SMITH, C. C.; FRETwELL, $. D. (1974): The optimal balance between size and number of offspring. Amer. Nat. 108, 499-506. Smurs, G. L. (1976): Population characteristics of Burchell’s zebra (Eguus burchelli antiquorum, H. Smith, 1841) in the Krueger National Park. S. Afr. J. Wildl. Res. 6, 99-112. SoKAL, R. R.; ROHLEF, F. J. (1969): Biometry. San Francisco: Freeman. SUNTZEFF, V.; CowDry, E. V.; Nıxon, B. B. (1962): Influence of maternal age on offspring in mice. J. Gerontol. 17, 2-7. TRIVERS, R. L. (1972): Parental investment and sexual selection. In: Sexual selection and the descent of man, 1871-1971. Ed. by B. G. CampgeLr. Chicago: Aldine. pp. 137-179. TRIVERS, R. L.; WILLARD, D. E. (1973): Natural selection of parental ability to vary the sex ratio of offspring. Science 179, 90-92. VAN DER MERWE, M.; SKINNER, J. D. (1982): Annual reproductive pattern in the dassie Procavia capensıs. S. Afr. J. Zool. 17, 130-135. VERME, L. J. (1983): Sex ratio varıation ın Odocoileus: a critical review. J. Wildl. Manage. 47, 573-582. — (1985): Progeny sex ratio relationships in deer: theoretical vs. observed. J. Wildl. Manage. 49, 134-136. VERME, L. J.; Ozoca, ]J. J. (1981): Sex ratio of white-tailed deer and the estrus cycle. J. Wildl. Manage. 45, 710-715. Vor, J. (1980): Generälni plemennäa kniıha koni Pfevalskeho —- General pedigree book of the Przewalski horse. Praha: Zoo Praha. — (1981-1984): Plemenna kniha koni Prevalskeho - Pedigree book of the Przewalski horse. Praha: 700; Praha. Warn, A.; WOLFOWITZ, J. (1940): On a test whether two samples are from the same population. Ann. math. Statist. 11, 147-162. WeELsH, D. A. (1975): Population, behavioural, and grazing ecology of the horses of Sable Island, Nova Scotia. Unpubl. PhD. thesis, Dalhousie Univ. WILLIAMS, G. C. (1979): The question of adaptive sex ratio in outcrossed vertebrates. Proc. roy. Soc., London, (B) 205, 567-580. WooLr, B. (1957): The log lıkelihood ratio test (the G-test). Methods and tables for tests of heterogeneity in contingency tables. Ann. Human Genet. 21, 397-409. ZABLOcKTJ, M. A. (1957): Nekotorye biologiceskie osobennosti zubra ı ich izmenenija v uslovijach zagonnogo soderZanija (Some biological peculiarities of the bison and their changes in captivity). ‚ Tr. prioksko-terrasnogo Zapoved. 1, 144-156. ZEGALOVv, $. B. (1950): Zakonomernosti nasledovanıja pola u Zivotnych (Regularities of the sex inheritance in anımals). Usp. sovr. Biol. 30, 130-144. Author’s address: Dr. Jırt MıikovskY, Department of Evolutionary Biology, Czechoslovak Academy of Sciences, Sekaninova 28, CS-120 00 Praha 2, Czechoslovakia Z. Säugetierkunde 53 (1988) 102-107 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 The karyotype of the European roe deer (Capreolus capreolus L.) By S. HERZOG Receipt of Ms. 26. 1. 1987 Abstract The karyotype of the European roe deer (Capreolus capreolus L.) ıs described by means of G- and C- banding techniques. A standardized idiogram (diagram, ordering of the chromosomes and designation of the bands) is proposed for the species Capreolus capreolus L., using the idiogram of the investigated Central European anımals as an example. Introduction Previous investigations on the karyotype of Capreolus capreolus L. used orcein staining (GusTavsson 1965; AMRUD and Nes 1966; HEerzoG and HöHn 1967; WURSTER and BENIRSCHRE 1967; GuSTAvssoN and SunDT 1968; Hsu and BENIRSCHKE 1968). A chromosone number ot 2n = 70, XX resp. XY was found. All 68 autosomes are designated as acrocentric, the X-chromosome submetacentric and the Y-chromosome subtelocentric. AMRUD and Nes (1966) call the Y “apparently submetacentric”. A supernumerary submetacentric chromosome was detected by HErzoG and HöHn (1967) in one specimen. The authors explained this phenomenon by a double trisomy, occurring as a centric fusion. More recently, banding techniques which allow to differenti- ate between single chromosomes were used by NEITZEL (1982) for the investigation of the Siberian roe deer (subspecies Capreolus capreolus pygargus). She studied three specimens and found a chromosome number between 2n= 76 and 80, dependent on a varying number of microchromosome pairs. Similar results have been obtained by SoKoLov et al. (1978), STUBBE and PassarGe (1979), and STUBBE (1979), using orcein staining techniques. In the European roe deer, a karyotype analysıs by means of modern banding techniques as well as a systematic description of the bands of each chromosome in accordance with an international standard is still lacking. The purpose of the present paper is to study the karyotype of the Central European roe deer by means of G- and C-banding techniques and to propose an idiogram, thereby describing the single chromosomes in accordance with the International System for Human Cytogenetic Nomenkclature (ISCN 1985), to set up a basis for further investigations, especially on homologisation of karyotypes between different taxa and studies concerning karyotype evolution. Material and methods In order to obtain the metaphase chromosomes, tissue cultures (kidney, skeletal muscle and testes) of 69 specimens from Hessen (Bundesrepublik Deutschland) were laid out using standard culture techniques. The metaphase chromosomes were studied by modified G- and C-banding techniques (SUMNER et al. 1971; SEABRIGHT 1972) and photographed with a 1000-fold magnification under oil immersion. In order to obtain a homogeneous degree of condensation in the G-banded chromosomes, only metaphases exhibiting nine G-bands on the X-chromosome (three on the short arm and sıx on the long arm) were taken into consideration. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5302-0102 $ 02.50/0 The era of Capreolus capreolus L. 103 31 2 33 34 Fig. 1. G-banded idiogram of a male European roe deer (Capreolus capreolus L.) Fıg. 2. Gonosomes (X and Y) of the European roe deer (Capreolus capreolus L.) 104 S. Herzog Table. Arm-length-ratio in the X-chromosome of Capreolus capreolus L., revealed by measure- ments of 26 haploid chromosome sets Arm relative length” (x) standard deviation (s,,) p-arm of! 0.30 q-arm 0.39 * as percentage of the total length of the haploıd autosome set. 123456789101 1213 14 15 16 17 18 19 202122 234252627 2829303 323334 X Y Fig. 3. Chromosome length in Capreolus capreolus L. (r. L. = relative length in percent of the total haploid autosome set) Results The somatic, diploiıd chromosome set of all 69 animals studied consists of 70 chromosomes, namely 68 autosomes and two gonosomes (2n = 70, XX resp. XY; figs. 1 and 2). Measurements of the chromosome lengths show that all autosome pairs and the Y- chromosome are telocentric (according to the terminology of NAacr 1980) with an arm- length ratio less than 1:4. The female gonosome is subtelocentric (see table). Using high- quality preparations, short (p-)Jarms are obvious ın all telocentric chromosomes. Thus, the Nombre fondamental (N. F.) ıs 140 ıf these short arms are taken into account, whereas it ıs 72 in the female and 71 in the male if the short p-arms are neglected. The autosomal N. F. (N. F.a) ıs 68 resp. 136 ın both male and female. The centromeric index (c.ı.) ıs not useful for the characterization of the chromosomes, because the p-arms of the telocentrics are not unequivocally measurable and chromosome lengths of the related chromosome pairs show only slıght differences (fig. 3), mostly less than the standard deviation of the chromosome length of the relevant pairs. Therefore it seems indispensable to use banding patterns as a tool for identification of the individual homologous chromosomes. Fig. 4 shows a proposal for the idiogram of Capreolus capreolus L. The bands and landmarks are drawn from the photos of the chromosomes of all 69 deer following the ISCN (1985). All autosomes exhibit distinet C- bands of different sıze. The euchromatın of a single autosome becomes stained over the whole length after C-banding treatment (facultative heterochromatin, fig. 5). The gonosomes show no visible C-bands, but in the female, one of the two X-chromosomes also exhibits facultative heterochromatin (sexchromatin), a phenomenon which is well known in other species, especially mammals. The occurrence of facultative heterochroma- (T snJ031dv9 snj024dvZ) 199p 301 ueadoing >feursf e Jo weıdorpı popueg-J ‘(1q31u) g 'Sı] - HWONU = 7 ‘spurq = sIaquunu [jfeuis ‘syrewpue] = siaqumu Stq :(G86T) NOSI ?YI SULMOJJOF "| S97094dv9 sn]094dvZ Fo weisorpt pxepueis e 107 uontsodosg ‘(1/27) # "317 43 HL N nm no N mn nm u 160 Konya um on nn N yenmsuone (@) (®) 77 [&e) — dl LEE um wm Non oo x 9 S = 4 1 7 2 8: l l : , 9) JE [33 [43 lE z € E 9 z [3 5 = I Sg (& ; 14 : E: z [4 I— 3 l [4 I E l (2 I [4 2 m EI 1 EI I ED 62 82 LE 9 E74 4 S 2 $ 9 [4 R 5 I 5, l : l s ä = 8 € E E E 1 ; ; Bes ol, $ l N N) u) 1 | Du N N I, I (5@ 44 le [074 61 8 7 14 7 5 E ; E E 5 i 2 5 £ a a E € m e 5 u f 2 l ? N z. l z l 2 | N IH 1 If ' Ol \ N EI NS SI Ne) = al Q -unmnm\SnV© MO Sr nm swonon KR -nmysoono an m = onm Da O [@) sn. 9m » Kun © O — - N = = I) S N nm wann Suno Ben on nenn YVanm“ a (©) [®) nn» N m mm N n Inu! SS. 0 h ©) n N (6) Em ee Sanenmven N oo ln nıı m (6) — m N TINTV in -Nmy non DD h (6) De N} [1] mn nnm no Duin o [@) m N mn >» MO (@) no N ae) a Na NINO KB © (6) u N {2%} NN Mm Mom MO = © — m N nenn m“ O oo 00) 106 S. Herzog tin in originally euchromatic regions is ınterpreted usually by the inactivation of one chromosome in a pair for gene dosis compensation (for a review see NA 1980). Discussion The present investigation on the karyotype of European roe deer confirms the results of Gustavsson (1965), HERZOG and HöHn (1967), WURSTER and BENIRSCHKE (1967), Gustavsson and SUNDT (1968) as well as Hsu and BENIRSCHKE (1968) as regards the chromosome number of Capreolus capreolus. The Siberian roe deer (Capreolus capreolus pygargus) also shows the characteristic chromosome number of 2n = 70 and, in addition, an unstable number of microchromosomes (SOKOLOV et al. 1978; STUBBE and PAssaRGE 1979; STUBBE 1979; NEITZEL 1982). NEITZEL (1982) assumes that the microchromosomes show regular mitotic segregation, because their number is constant within each investi- gated individual. A supernumerary submetacentrice chromosome detected by HERZOG and Hörn (1967) in one specimen of roe deer from Hessen (Bundesrepublik Deutschland) could not be found in the present investigation nor is ıt mentioned by other authors, which means that such a chromosome mutation is either uncommon in the European roe deer or it is lethal at an early ontogenetic stage. With respect to the form of the chromosomes, the previous investigations could be confirmed only partially, because the short p-arms are not described in these papers. In consequence, the autosomes are not acro- but telocentric, according to the terminology given by Nacı (1980). Moreover, the Y-chromosome is also telocentric and not subtelocentric or submetacentric as assumed by previous authors. The X-chromosome was identified as subtelocentric. The photos of the orcein stained chromosomes given in the above mentioned first papers about the karyotype of roe deer suggest that the differences in the morphology of the gonosomes, especially the Y- chromosome, might be explainable by the unconsistent use of the terminology. However, it also may be possible that Y chromosomes with p-arms longer than those of the autosomes are existing, but there is no certain indication. Further investigations on this question should be carried out using electron microscopy. Photos of G-banded chromosomes 1 to 16 are shown by NEITZEL (1982) for the Siberian roe deer (Capreolus capreolus pygargus). Although these chromosomes seem to be stained at amore condensed stage, the landmark patterns are similar to those of the investigated animals from Hessen. This indicates that there might be no considerable differences between the banding patterns of chromosomes 1 to 16 ot the deer from Central Europe and from Sıberia. Moreover, the other autosomes also seem to be similar ın the Sıberian and the Central European roe deer as regards their G-banding pattern, which is described by NEITZEL (1982), who compares the karyotypes of different Cervidae, including the Siberian roe deer, with a karyotype assumed to be ancestral for the Cervidae. The C-banding patterns described by NEITZEL (1982) are sımilar to those revealed ın the present investigation: She also found C-bands in each autosome, but not distinct C-bands in the gonosomes. The microchromosomes of the Siberian roe deer are described as fully heterochromatic. Banding techniques, such as G-banding, and a standard basis for description of the single chromosomes, such as the ISCN (1985), should enable the investigator to identify nearly all chromosomes reliably and appear to be useful for homologisation of the karyotypes between different taxonomic groups as well as for studies on karyotype evolution or for gene mapping. Moreover, C-banding is very important, especially for investigations on karyotype evolution. For studies on population genetics in this species, C-banding may become useful if any C-band polymorphisms can be found and identified as markers by genetic analysis. The karyotype of Capreolus capreolus L. 107 Acknowledgements The author is very grateful to Prof. Dr. A. Herzoc and Dr. Hennı Höhn, Fachgebiet Veterinär- medizinische Genetik und Zytogenetik, Institut für Tierzucht und Haustiergenetik der Universität Gießen, for the help and advice during the experimental work. The studies were supported financially by the Hessischer Minister für Landwirtschaft und Umwelt. Zusammenfassung Der Karyotyp des europäischen Rehes (Capreolus capreolus L.) Der Karyotyp des mitteleuropäischen Rehes wurde mittels G- und C-Bandentechnik beschrieben. Ein standardisiertes Idiogramm (Anordnung der Chromosomen und Bezeichnung der Banden) wurde für die Species Capreolus capreolus L. am Beispiel des Idiogramms der untersuchten mitteleuropäischen Tiere vorgeschlagen. References Standing Committee on Human Cytogenetic Nomenclature (1985): An International System for Human Cytogenetic Nomenclature (1985). Basel: Karger. AMRUD, J.; Nes, N. (1966): The chromosomes of the roe (Capreolus capreolus). Hereditas 56, 217-220. GusTavsson, 1. (1965): Chromosome studies in five species of deer representing the four genera Alces, Capreolus, Cervus and Dama. Mamm. Chromos. Newslett. 18, 149. GusSTAVSSoN, 1.; SUNDT, C. ©. (1968): Karyotypes ın 5 species of deer (Alces alces L., Capreolus capreolus L., Cervus elaphus L., Cervus n. nippon Temm. and Dama dama L.). Hereditas 60, 233-248. HERZoc, A.; Hönn, H. (1967): Darstellung der Chromosomen aus Knochenmarkszellen beim Reh, Capreolus capreolus, und Rottier, Cervus elaphus. Z. Jagdwiss. 13, 118-121. Hsu, T. C.; BENIRSCHKE, K. (1968): An atlas of mammalıan chromosomes. Vol. 2, Folio 41/42/43. New York: Springer. Nacrı, W. (1980): Chromosomen. Berlin und Hamburg: Paul Parey. NEITZEL, H. (1982): Karyotypenevolution und deren Bedeutung für den Speziationsprozefß der Cerviden (Cervidae; Artiodactyla; Mammalıia). Berlin: Diss. thesis, Freie Universität Berlin. SEABRIGHT, M. (1972): The use of proteolytic enzymes for the mapping of structural rearrangements in the chromosomes of man. Chromosoma 36, 204-210. SoKoLov, V. E.; OrLow, V. N.; CHUDINOVSKAYA, G. A.; DAanILkın, A. A. (1978): Differences in chromosomes between two subspecies, Capreolus capreolus capreolus L. and C. c. pygargus Pall. Zool. Zh. 57, 1109-1112. STUBBE, C. (1979): Bastardierungsversuche zwischen Europäischen und Sibirischen Rehen. Wissen- schaft und Fortschritt 29, 113-116. STUBBE, C.; PASSARGE, H. (1979): Rehwild. Melsungen: Neumann-Neudamm. SUMNER, A. T.; Evans, H. ]J.; BuckLAnD, R. A. (1971): New technique for distinguishing between human chromosomes. Nature New Biol. 232, 31-32. WURSTER, D. H.; BENIRSCHKE, K. (1967): The chromosomes of the Cervoıdea and Bovoidea. Mamm. Chromos. Newslett. 8, 226-229. Author’s address: SVEN HERZOG, Abteilung Forstgenetik und Forstpflanzenzüchtung, Universität Göttingen, Büsgenweg 2, D-3400 Göttingen Z. Säugetierkunde 53 (1988) 108-123 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Osteomorphological features of the appendicular skeleton of African buffalo, Syncerus caffer (Sparrman, 1779) and of domestic cattle, Bos primigenius f. taurus Bojanus, 1827 By J. PETERS Laboratorium voor Paleontologie, Rijksuniversiteit Gent, Gent, Belgium Receipt of Ms. 10. 11. 1986 Abstract Studied the osteomorphological differences between the appendicular skeleton of African buffalo (Syncerus caffer) and domestic cattle (Bos primigenins f. taurus). Osseous remains derived from these large bovids, frequently found in African Holocene archaeological sites, can not be distinguished easily. A key has been developed to meet this recurrent problem and a number of diagnostic, osteomor- phological features are established, which allow a distinction between the two species. Only a few of the smaller carpal and tarsal bones can not be separated yet. In general, osteomorphological differences are more constant than osteometrical differences and therefore seem more useful. Most of the ostemorphological criteria, established for domestic cattle can also be used to identify remains of their wild ancestor, the aurochs (Bos primigenius). Introduction The following study was undertaken within the frame of our Ph. D. research on faunal remains from archaeological sites in Central and Eastern Sudan (cf. Marks et al. 1985; PETERS 1986a, 1986b). During this archaeozoological analysıs, we were confronted with the fact that the majority of our samples was dominated by osseous remains from members of the family Bovidae, ranging in size from the small orıbi (Ourebia onrebi) up to the large buffalo (Syncerus caffer). Because of (1) the diversity of bovid species within these collections (up to 20 species or more), (2) their mixed composition wıth domesticated and wild bovids and (3) the pronounced fragmentation of the bone material, their ıdentification presented considerable problems. The literature available on African bovid osteology focuses mainly on the morphology of the skull, including the teeth (e.g. ARAMBOURG 1947; GENTRY 1964, 1967, 1978; STÖCKMANN 1975; Van NEER 1981 and others). Postcra- nıal skeletons, however, are poorly known, for descriptions of their osteomorphological characteristics, useful to the archaeozoologist, are quite rare (ARAMBOURG 1947; GENTRY 1967; LEINDERS and SONDAAR 1974; OBOoussIER and Ernst 1977; LEINDERS 1979; VAN NEER 1981; GABLER 1985; WALKER 1985). To solve partly our identification problems, we carried out a few osteomorphological studies on recent and fossil postceranial material of African and other bovids. The choice of the species considered in these contributions is conditioned by an important question concerning the life style of prehistoric man: are domesticated animals present ın our collections or not? Therefore, this first analysis deals with the osteomorphology of two very large bovids, of which, until now, the postcranial skeleton could not be separated accurately: the Afrıcan buffalo, Syncerus caffer and domestic cattle, Bos primigenius f. taurus. Within the descriptive part, we include several distinctive features already recorded by other authors ın earlier publications (DOTTRENS 1946; GENTRY 1967) or reports (PAYNE s.d.). To distinguish between the phalanges of the fore and hind limbs of cattle, we used U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5302-0108 $ 02.50/0 The appendicular skeleton of African buffalo and of domestic cattle 109 some of the criteria established by DoTTREns (1946). As to the work by GEnTRY (1967), we do not agree with the conclusions concerning the distinction between certain skeletal elements of Bos and Syncerus. We suspect that the small size of the sample used by this researcher may be responsible for our differences of opinion. In the course of our study, we also collected an impressive amount of osteometrical data, which enabled us to calculate many indices. This information has not been included here for practical reasons, but ıt can be obtained from the author at the address listed below. Both these osteometrical data and the ones summarized here will be available soon in an extensive, technical paper (PETERS 1986c). This paper is distributed on a very limited scale; therefore we thought it useful to publish separately the following short article. Material and methods The following results are based on a detailed analysis of the appendicular skeleton of the two species involved. As to the African buffalo (Syncerus caffer), 25 adults, including both sexes, were carefully examined. All three subspecies sensu HALTENORTH and DiILLEr (1979:95) are present: the forest buffalo ($. c. nanus), the western savanna buffalo ($. c. brachyceros) and the savanna buffalo ($. c. caffer). The specimens studied are collected from all over Africa, but mainly Zaire. They are stored in the Koninklijk Museum voor Midden-Afrika, Tervuren-Belgium; the Koninklijk Belgisch Instituut voor Natuurwetenschappen, Brussels and the British Museum (Natural History), London. From cattle (Bos primigenius f. taurus), 15 adults, including both sexes and hundreds of fossil specimens collected in archaeological sites of varyıng ages in Europe (Neolithic to Modern Times) were examined. The recent material consists of European as well as African specimens of extant breeds. This material is stored in the institutions already mentioned, and partly in the Laboratorium voor Paleontologie and the Laboratorium voor Anatomie van de Huisdieren, both at the Rijksuniver- siteit Gent. For the osteomorphological descriptions, we have followed strictly the nomenclature proposed by the International Committee on Veterinary Gross Anatomical Nomenclature in their “Nomina Anatomica Veterinaria’ (3rd. ed., 1983). The figures were drawn by Mrs. J. BAETEns from right limb bones with the light coming from the lefthand top corner; each scale bar represents 20 mm. Note that the first and second phalanges belong to the fourth digit; the third phalanges are taken from the third digit. We did not consider the dew claws ın this study. Results Osteomorphological features of the appendicular skeleton of African buffalo and cattle The relevant dıiagnostic features are indicated by a number between brackets, which is also given on the plates. Arrows on these plates indicate morphological differences, lines refer to general differences in proportions. Scapula 1. The position of the spina scapulae differs in the two genera. In Bos, the spina scapulae is slightly curved so that the acromion projects across the line of the margo cranialis when the bone is laterally viewed (pl. 1, fig. 1, char. 1). In Syncerus the ventral portion of the spina scapulae appears to be rather straight, so that the acromıon remains within the line of the margo cranıalıs (pl. 1, fig. 2). As a consequence, the width ratio fossa supraspinata: fossa infraspinata is circa 1 to 3 in Bos, in stead of 1 to 2 or 2.5 in Syncerus. 2. The lateral border of the cavitas glenoidalıs exhibits a medial notch in Bos (pl. 1, fig. 3, char. 2). In Syncerus, a comparable notch has been observed only once; in all other specimens ıt was less pronounced or even absent (pl. 1, fig. 4). 3. In Syncerus, the incısura glenoidalis is well developed, while in Bos it is almost completely absent (pl. 1, figs. 3-4, char. 3). 12 Jo Baelens Plate 1. 1: Scapula, lateral view, Bos primigenius f. taurus, 2: Scapula, lateral view, Syncerus caffer, 3: Scapula, distal view, Bos primigenius f. taurus, 4: Scapula, distal view, Syncerus caffer, 5: Humerus, proximal extremity, cranial view, Bos primigenins f. taurus, 6: Humerus, proximal extremity, cranial view, Syncerus caffer, 7: Humerus, proximal extremity, lateral view, Bos primigenius f. taurus, 8: Humerus, proximal extremity, lateral view, Syncerus caffer, 9: Humerus, distal extremity, lateral view, Bos primigenius f. taurus, 10: Humerus, distal extremity, lateral view, Syncerus caffer, 11: Humerus, distal extremity, medial view, Bos primigenins f. taurus, 12: Humerus, distal extremity, medial view, Syncerus caffer Jo Daclens 2 (10) Plate 2. 1: Radius-Ulna, proximal extremity, proximal view, Bos primigenins f. taurus, 2: Radıus-Ulna, proximal extremity, proximal view, Syncerus caffer, 3: Ulna, olecranon, lateral view, Bos primigenius f. taurus, 4: Ulna, olecranon, lateral view, Syncerus caffer, 5: Radius-Ulna, distal extremity, cranıal view, Bos primigenius f. taurus, 6: Radıus-Ulna, distal extremity, cranial view, Syncerus caffer, 7: Os femoris, proximal extremity, caudal view, Bos primigenius f. taurus, 8: Os femoris, proximal extremity, caudal view, Syncerus caffer, 9: Os femoris, proximal extremity, cranıal view, Bos primigenins f. taurus, 10: Os femoris, proximal extremity, cranial view, Syncerus caffer, 11: Os femoris, distal extremity, caudal view, Bos primigenius f. taurus, 12: Os femoris, distal extremity, caudal view, Syncerus caffer ul J. Peters Humerus 1. The position of the pars caudalıs of the tuberculum majus, relative to that of the pars cranıalis differs in the two genera. In a cranıal view, the pars caudalis projects more laterally compared with the pars cranialis in Bos, while in Syncerus both are lying more or less in the same plane (pl. 1, fıgs. 5-6, char. 4). 2. The pars caudaliıs of the tuberculum majus ıs proximally and caudally more developed ın Bos compared wıth Syncerus (pl. 1, figs. 7-8, char. 5) (see also GENTRY 1967:284-char. 71). 3. A lateral view of the humerus of Bos shows that the facies musculi infraspinatı is well developed cranially, through which it forms a projection at the cranıal side of the humerus. In Syncerus, this rough prominence is less pronounced and less well developed cranially (pl. 1, fıgs. 7-8, char. 6) (see also GENTRY 1967: 284-char. 72). 4. The transıtion between the epicondylus lateralis humeri and the fossa radialıs humeri isn Bos characterızed by a cranioproxımal, rather pointed attachment surface. In Syncerus, this attachment area is less pronounced (pl. 1, figs. 9-10, char. 7). 5. The epicondylus medialis is more developed distally in Bos compared with Syncerus (Pla hiesall 12, 2char 8): Radius 1. The margo caudalıs of the proximal artıcular surface shows a different course in both genera. This is due to the differences in form and proportions of the lateral part of the incisura ulnaris (pl. 2, figs. 1-2, char. 9). 2. The portion of the margo cranıalıs of the facıes artıcularıs carpea, which corresponds with the dorsal border of the os carpı intermedium, extends more distally in Syncerus (pl. 2, figs. 5-6, char. 10). Ulna 1. In Bos, the processus coronoideus lateralis is decidedly more developed laterally compared with Syncerus (pl. 2, figs. 1-2, char. 11). 2. In Bos, the incısura lateralis has a rectangular form, while in Syncerus this incisura is rather trıangular and less well pronounced at both its dorsal and lateral side (pl. 2, figs. 1-2, char. 12). 3. The tuber olecrani exhibits ın Bos a distinct proximal notch which ıs almost lacking ın Syncerus (pl. 2, figs. 3-4, char. 13). Ossa carpı Os carpı radiale. 1. The ratio of the proximodistal versus dorsopalmar dimensions is different in the two genera (pl. 4, figs. 1-2, char. 14). 2. The margo medıalıs exhibits a slightly more angular course in Bos in comparison with Syncerus (pl. 4, figs. 3-4, char. 15) (see also GENTRY, 1967: 284-char. 83). Os carpi intermedium. 1. The margo palmaris of the facies articularıs proximalis is more developed proximally in Bos (pl. 4, figs. 5-6, char. 16). 2. The angle between the palmar border and the (oblique) medial border of the facies articularıs distalis is about 45° ın Syncerus, while in Bos this angle ıs about 30° (pl. 4, figs. 5-6, char. 17). Os carpi ulnare. The facies articularis medialis of the os carpi ulnare is in Bos much more pronounced in comparison with Syncerus (pl. 4, fıgs. 7-10, char. 18). Os carpı accessorıum. No constant osteomorphological differences were found. The appendicular skeleton of African buffalo and of domestic cattle 15 Os carpale II + III. 1. In a proximal view, the habitus of the os carpale II + III is rather squarish in Syncerus, while in Bos this carpal bone looks more rectangular because of an increased mediolateral distance (pl. 4, figs. 11-12, char. 19). 2. In Bos, the medial articular surface is cut into two parts by a distopalmar groove. In Syncerus, this medial artıcular surface remains uniform (pl. 4, figs. 13-14, char. 20). Os carpale IV. No constant osteomorphological differences were found. Os metacarpale III + IV 1. The habitus of the os metacarpale III + IV differs in the two genera: relatively slender in Bos, while shorter, broader and rather sturdy in Syncerus (pl. 4, figs. 15-16, char. 21) (see partly GENTRy 1967: 282-char. 62). 2. The foramen nutricium at the palmar side of the distal extremity is well developed in Bos, while in Syncerus this foramen is reduced or even absent (pl. 4, figs. 15-16, char. 22) (see also GENTRY 1967: 282-char. 66). 3. The tuberositas ossis metacarpalis III is more pronounced in Bos than in Synceruns (pl. 4, figs. 17-18, char. 23). Os femoris 1. The central portion of the crista intertrochanterica has a minor mediodorsal fold, which is absent in Syncerus (pl. 2, figs. 7-8, char. 24). 2. The caput ossis femoris merges gradually into the trochanter major in Bos, while in Syncerus the edge of the caput ossis femoris forms a clear boundary between the medial and lateral parts of the proximal extremity (pl. 2, figs. 7-8, char. 25). We agree wıth GENTRY (1967: 280-char. 49) that Bos tends to have a steeper slope on the top edge ot the artıcular head in anterior view compared with Syncerus, although this feature is not distinguishable in every bone or bone fragment. 3. In Syncerus, a foramen nutricium is present near the proximal end of the femur. In Bos, a comparable foramen is located at the caudal side of the femur diaphysis near the distal end, slightly proximomedial of the fossa supracondylarıs (pl. 2, figs. 9-12, char. 26). 4. The medial ridge of the trochlea ossis femoris extends more proximally in Bos; this trochlea is altogether more developed proximally compared with its analogue in Syncerus (pl. 3, figs. 1-2, char. 27). 5. The lateral ridge of the trochlea ossis femoris is more pronounced distally in Syncerus (pl. 3, fıgs. 3-4, char. 28). Patella The patella of Bos generally has, in comparison wıth Syncerus, a more slender habitus; this is partly due to a prolonged proximodistal axis (pl. 3, figs. 5-6, char. 29). Tıbia The sulcus malleolaris lateralis is more pronounced in Bos. The morphology of the facies artıcularıs malleolı is also different in the two genera (pl. 3, figs. 7-10, char. 30). Os malleolare The cranıoproximal portion of the os malleolare of Syncerus is ın most cases protruding proximally (pl. 4, figs. 11-12, char. 31). > { Da Plate 3. 1: Os femoris, distal extremity, medial view, Bos primigenius f. taurus, 2: Os femoris, distal extremity, medial view, Syncerus caffer, 3: Os femoris, distal extremity, lateral view, Bos primigenins f. taurus, 4: Os femoris, distal extremity, lateral view, Syncerus caffer, 5: Patella, caudal view, Bos primigenius f. taurus, 6: Patella, caudal view, Syncerus caffer, 7: Tibıa, distal epiphysis, distal view, Bos primigenius f. taurus, 8: Tıbia, distal epiphysis, distal view, Syncerus caffer, 9: Tibia, distal extremity, lateral view, Bos primigenius f. taurus, 10: Tibia, distal extremity, lateral view, Syncerus caffer, 11: Os malleolare, lateral view, Bos primigenius f. taurus, 12: Os malleolare, lateral view, Syncerus caffer Jo Baetens Plate 4. 1: Os carpı radiale, dorsomedial view, Bos primigenius f. taurus, 2: Os carpı radiale, dorsomedial view, Syncerus caffer, 3: Os carpı radiale, proximal view, Bos primigenins f. taurus, 4: Os carpı radiale, proximal view, Syncerus caffer, 5: Os carpı intermedium, proximal view, Bos primigenius f. taurus, 6: Os carpi intermedium, proximal view, Syncerus caffer, 7: Os carpi ulnare, dorsal view, Bos primigenins t. taurus, 8: Os carpı ulnare, dorsal view, Syncerus caffer, 9: Os carpı ulnare, proximal view, Bos primigenins f. taurus, 10: Os carpı ulnare, proximal view, Syncerus caffer, 11: Os carpale II + III, proximal view, Bos primigenius f. taurus, 12: Os carpale II + III, proximal view, Syncerus caffer, 13: Os carpale II + III, medial view, Bos primigenins f. taurus, 14: Os carpale II + III, medial view, Syncerus caffer, 15: Os metacarpale III + IV, palmar view, Bos primigenins f. taurus, 16: Os metacarpale III + IV, palmar view, Syncerus caffer, 17: Os metacarpale III + IV, proximal epiphysis, proximal view, Bos primigenins f. taurus, 18: Os metacarpale III + IV, proximal epiphysis, proximal view, Syncerus caffer Plate 5. 1: Talus, plantar view, Bos primigenius f. taurus, 2: Talus, plantar view, Syncerus caffer, 3: Calcaneus, plantar view, Bos primigenins f. taurus, 4: Calcaneus, plantar view, Syncerus caffer, 5: Calcaneus, medial view, Bos primigenins f. taurus, 6: Calcaneus, medial view, Syncerus caffer, 7: Os centroquartale, lateral view, Bos primigenius f. taurus, 8: Os centroquartale, lateral view, Syncerus caffer, 9: Os centroquartale, proximal view, Bos primigenins f. taurus, 10: Os centroquartale, proximal view, Syncerus caffer, 11: Os centroquartale, distal view, Bos primigenins f. taurus, 12: Os centroquartale, distal view, Syncerus caffer, 13: Os tarsale II + III, proximal view, Bos primigenins f. taurus, 14: Os tarsale II + III, proximal view, Syncerus caffer, 15: Os metatarsale III + IV, proximal epiphysis, proximal view, Bos primigenius f. taurus, 16: Os metatarsale III + IV, proximal epiphysis, proximal view, Syncerus caffer, 17: Os metatarsale III + IV, dorsal view, Bos primigenius f. taurus, 18: Os metatarsale III + IV, dorsal view, Syncerus caffer Jo Baetens (47) Plate 6. 1: P. proximalis manus, abaxial view, Bos primigenins f. taurus, 2: P. proximalis manus, abaxial view, Syncerus caffer, 3: P. proximalıs pedis, abaxial view, Bos primigenius f. taurus, 4: P. proximaliıs pedis, abaxial view, Syncerus caffer, 5: P. proximalıs manus, axial view, Bos primigenius f. taurus, 6: P. proximalis manus, axial view, Syncerus caffer, 7: proximalıs pedis, axial view, Bos primigenius f. taurus, 8: P. proximalis pedis, axial view, Syncerus caffer, 9: P. proximalis manus (2), proximal view, Bos primigenins f. taurus, 10: P. proximalıs manus (2), proximal view, Syncerus caffer, 11: P. proximalis manus (d), proximal view, Bos primigenins f. taurus, 12: P. proximalis manus (d), proximal view, Syncerus caffer, 13: P. proximalıs pedis (2), proximal view, Bos primigenius f. taurus, 14: P. proximalis pedis (?), proximal view, Syncerus caffer, 15: P. proximalıs pedis (), proximal view, Bos primigenins f. taurus, 16: P. proximalis pedis (JS), proximal view, Syncerus caffer (51) Jo Baelens Plate 7. 1: P. media manus, abaxial view, Bos primigenins f. taurus, 2: P. media manus, abaxial view, Syncerus caffer, 3: P. media pedis, abaxial view, Bos primigenins f. taurus, 4: P. media pedis, abaxial view, Syncerus caffer, 5: P. media manus, palmar view, Bos primigenius f. taurus, 6: P. media manus, palmar view, Syncerus caffer, 7: P. media pedis, plantar view, Bos primigenins f. taurus, 8: P. media pedis, plantar view, Syncerus caffer, 9: P. media manus (2), proximal view, Bos primigenius f. taurus, 10: P. media manus (2), proximal view, Syncerus caffer, 11: P. media manus (d), proximal view, Bos primigenius 1. taurus, 12: P. media manus (Ö), proximal view, Syncerus caffer, 13: P. media pedis (?), proximal view, Bos primigenius f. taurus, 14: P. media pedis (2), proximal view, Syncerus caffer, 15: P. media pedis (JS), proximal view, Bos primigenius f. taurus, 16: P. media pedis (S), proximal view, Syncerus caffer Jo Baetens 2 kon re DRISST ee 9 ATONAT | Rena fi Ay), N J hi. PR $ ee 27 ) FREIE Fr F—_ „)]: Plate 8. 1: P. distalis manus, abaxıal view, Bos primigenius f. taurus, 2: P. distalıs manus, abaxıal view, Syncerus caffer, 3: P. distalis pedis, abaxial view, Bos primigenins f. taurus, 4: P. distalis pedis, abaxiıal view, Syncerus caffer, 5: P. distalis manus, axial view, Bos primigenins f. taurus, 6: P. distalıs manus, axial view, Syncerus caffer, 7: P. distalis pedis, axial view, Bos primigenius f. taurus, 8: P. distalis pedis, axial view, Syncerus caffer, 9: P. distalis manus, proximal view, Bos primigenins f. taurus, 10: P. distalis manus, proximal view, Syncerus caffer, 11: P. distalis pedis, proximal view, Bos primigenius f. taurus, 12: P. distalis pedis, proximal view, Syncerus caffer 120 J. Peters OÖssa tarsı Talus. In many cases, the caput talı exhibits in Bos at its facies articularıs ossis centroquar- talıs a lateral groove, which is absent ın Syncerus (pl. 5, fıgs. 1-2, char. 32). Calcaneus. 1. In Syncerus, the sustentaculum tali is more pronounced medially (pl. 5, fig. 4) while ın Bos ıt is more developed ın a plantar direction (pl. 5, fig. 5, char. 33). 2. The proximal portion of the processus coracoideus is better developed dorsally in Bos; the transition towards the proximal part of the calcaneus lies more plantarly compared with Syncerus (pl. 5, fig. 5, char. 34). Os centroquartale. 1. The plantar side of the lateral half of the os centroquartale exhibits in Bos a well pronounced plantar prominence, which is nearly absent in Syncerus (pl. 5, figs. 7-8, char. 35). 2. In Bos, the medioplantar portion of the proximal articular surface of the os centroquartale, which articulates with the caput tali, shows an extra articular surface laterally (pl. 5, figs. 9-10, char. 36). 3. The small, distal, lateroplantar articular surface, which articulates wıth a corresponding surface at the proximal extremity of the os metatarsale III + IV is ın Bos generally smaller than in Syncerus (pl. 5, figs. 11-12, char. 37). Os tarsale I. No constant osteomorphological differences were found. Os tarsale II + III. No constant osteomorphological differences were found. Os metatarsale III+IV 1. The habitus of the os metatarsale III+IV differs in the two genera: relatively slender in Bos, while shorter, broader and rather sturdy in Syncerus (pl. 5, fıgs. 17-18, char. 38). 2. The lateroplantar articular surface of the proximal epiphysis is much more developed laterally in Syncerus (pl. 5, figs. 15-16, char. 39). Ossa digitorum Criteria to distinguish the ossa digitorum manus from the ossa digitorum pedis ın Bos and Syncerus Phalanges proximales. 1. The habitus of the P. proximales pedis is more slender compared with that of the P. proximales manus (pl. 6, figs. 1-8, char. 40) (see also DOTTREns, 1946:764). 2. The general appearance of the proximal end of the first phalanges is rather squarısh for those of the fore limb and rather rectangular for those of the hind limb (pl. 6, fıgs. 9-16, char. 41) (see also DOTTRENS 1946:765). 3. In Bos, the artıcular surface for the axial os sesamoideum proximale of the P. proximales manus is reduced in size compared with that of the P. proximales pedis (pl. 6, figs. 9-16, char. 42) (see also DoTTREns 1946:765). Phalanges mediae. 1. The habitus of the P. mediae pedis of Bos and Syncerus ıs more slender compared with that of the P. mediae manus (pl. 7, figs. 1-8, char. 43) (see also DOTTREns 1946:753). 2. The general appearance of the proximal end of the phalanges mediae is rather squarish for those of the fore limb, and rather rectangular for those of the hind lımb (pl. 7, fiıgs. 9-16, char. 44). 3. In Bos, the abaxiopalmar part of the trochlea phalangis mediae manus is more developed proximally compared with its analogue in the P. mediae pedis (pl. 7, fıgs. 5 and 7, char. 44a) (see also DOTTRENS, 1946:753). The appendicular skeleton of African buffalo and of domestic cattle 121 Phalanges distales. In axial view, it becomes obvious that the margo coronalis of the distal phalanges of the hind limb exhibits a steeper course than that of the distal phalanges of the fore limb (pl. 8, fıgs. 5-8, char. 45) (see also DOTTRENS 1946:743). Criteria to distinguish between the ossa digitorum from Bos and Syncerus Phalanges proximales. 1. In both axial and abaxial view, one notices the angular aspect of the phalanges proximales in Bos, while in Syncerus these phalanges are more rounded (pl. 6, figs. 1-8, char. 46). We nevertheless agree with $. Payne (in litt.) that this criterium cannot always be used. 2. The proximal fovea articularıs is well delineated in Syncerus, which is not the case in Bos (pl. 6, figs. 9-16, char. 47). 3. The facies articulares for the ossa sesamoidea proximalia are more pronounced in Syncerus compared with Bos (pl. 6, figs. 9-16, char. 48). Phalanges mediae. 1. In Syncerus, the phalanges mediae generally show a more slender habitus compared with those from Bos (pl. 7, figs. 1-8, char. 49). 2. In Syncerus, the abaxiopalmar part of the trochlea phalangis mediae manus is less developed proximally compared with its analogue in Bos (pl. 7, fıgs. 5-6, char. 50). 3. The articular surface is divided into two glenoid cavities by a crista sagittalis. In Bos, the difference in size between the abaxial and axial glenoid cavities is much larger compared with Syncerus (pl. 7, figs. 9-16, char. 51). 4. In many cases, the abaxial tuberosity of the torus palmaris/plantaris is less pronounced in Bos (pl. 7, figs. 9-16, char. 51a) (see also Payne, unpublished report). Phalanges distales. 1. The processus extensorius is more developed in Bos (pl. 8, figs. 1-8, char. 52). 2. The tuberculum flexorium is in Bos more pronounced in the palmar (P. distales manus) and plantar (P. distales pedis) direction (pl. 8, figs. 1-8, char. 53). 3. The facıes articularis sesamoidea for the os sesamoideum distale is larger and lies more plantarly in Bos (pl. 8, fıgs. 9-12, char. 54). 4. In Bos, the axial border of the facies articularis is indented, which is not the case in Syncerus (pl. 8, figs. 9-12, char. 55). Concluding remarks From the foregoing, ıt should be clear that a number of diagnostic osteomorphological features exist which allow a distinction between African buffalo and cattle. Only a few smaller carpal and tarsal bones such as the os carpi accessorium, the os carpale IV, the os tarsale I and the os tarsale II+III cannot be separated yet morphologically. Due to the fact that many features are located near the articular surfaces of the bones, even incomplete bones - in casu fossil specimens — can now in many cases be identified to the species level. During our analysıs, we also found out that measurements, and the indices based on them, proved to be a less useful tool for the distinction between the skeletal elements of the two species, because of the large overlap. We furthermore were able to check whether the osteomorphological characteristics, established for domestic cattle, were also applicable to ıts wild ancestor, the aurochs (Bos primigenius). It ıs known that the domestication process causes morphological changes but, from our observations, we can conclude that most of the features of domestic cattle described above can also be used to identity its wıld ancestor. Acknowledgements The author is indebted to Drs. A. GAUTIER, P. SIMOENS, Rijksuniversiteit Gent, and $. Payne, Cambridge University, for reading the manuscript and discussing the subject; to Drs. W. van NEER, Katholieke Universiteit Leuven, and J.-P. BRUGAL, C.N.R.S., Marseille, for their valuable comments; 122 J. Peters to Drs. X. Mısonne, Koninklijk Belgisch Instituut voor Natuurwetenschappen, Brussels, D. Thıjs VAN DEN AUDENAERDE, D. MEIRTE, Koninklijk Museum voor Midden-Afrika, Tervuren, and ]. CruTTon-Brock and K. Bryan, British Museum Natural History, London, for the permission to study museum material; to J. BAETEns for the drawings and to N. REYNAERT for typing the manuscript. This study has been financed by the I.W.O.N.L., Brussels; a travel grant was provided by the Vlaamse Wetenschappelijke Stichting, Leuven. Zusammenfassung Osteomorphologische Unterscheidungsmerkmale am Gliedmaßenskelett vom afrıkanischen Büffel (Syncerus caffer) und vom Hausrind (Bos primigenins f. tanrus) Knochenresten von diesen großen Boviden werden oft gefunden an afrıkanischen holozänen archaeo- logischen Fundorten, aber ihre Bestimmung schafft manches Problem. Ein Bestimmungsschlüssel wurde entwickelt, um dieses immer wiederkehrende Problem zu lösen; die diagnostischen, osteomorphologischen Merkmale, welche eine Unterscheidung beider Tierarten voneinander ermöglichen, werden festgelegt. Nur einige kleine Karpal- und Tarsalknochen können noch nicht unterschieden werden. Im allgemeinen sind die osteomorphologischen Unterscheidungsmerkmale beständiger als die osteometrischen. Den größeren Teil dieser osteomorphologischen Charakteristiken, festgelegt für das Hausrind, kann man auch anwenden, um Knochenreste ihres Vorfahren, des Ur, zu bestimmen. References ARAMBOURG, C. (1947): Contribution & l’etude geologique et paleontologique du Bassin du Lac Rodolphe et de la Basse Vallee de l’Omo. 2° partie: Pale&ontologie. Mission scientifique de ’Omo 1932-1933, Tome I (Geologie et Anthropologie), Fascicule III. Ed. Mus. Nat. d’Hist. Nat. Parıs. BoEssnEcK, J.; MÜLLER, H.-H.; TEICHERT, M. (1964): Osteologische Unterscheidungsmerkmale zwischen Schaf (Ovis aries Linne) und Ziege (Capra hircus Linne). Kühn-Archiv 78, 1-129. Dortkens, E. (1946): 1. Etude preliminaire: Les phalanges osseuses de Bos Taurus domesticus. Rev. Suisse de Zool. 53(33), 739-774. GABLER, K.-O. (1985): Osteologische Unterscheidungsmerkmale am postkranialen Skelett zwischen Mähnenspringer (Ammotragus lervia), Hausschaf (Ovis aries) und Hausziege (Capra hircus). München: Diss.- u. Fotodruck Frank GmbH. GENTRY, A. W. (1964): Skull characters of African gazelles. Ann. and Mag. Nat. Hist. Ser. 13, 7, 353-382. — (1967): Pelorovis oldowayensis RECK, an extinct bovid from East Africa. Bull. Brit. Mus. (Nat. Hist.), Geol. 14(7). — (1978): Bovidae. In: Evolution of African Mammals. Ed. by Macıio, V. J.; Cook£, H. B. S. Cambridge, Mass. and London: Harvard Univ. Press. 539-572. HALTENORTH, TH.; DILLER, H. (1979): Elseviers Gids van de Afrikaanse zoogdieren. Amsterdam and Brussel: Elsevier. LEINDERS, J. J. M. (1979): On the osteology and function of the digits of some ruminants and their bearing on taxonomy. Z. Säugetierkunde 44, 305-319. LEINDERS, J. J. M.; SONDAAR, P. Y. (1974): On functional fusions in footbones of Ungulates. Z. Säugetierkunde 39, 109-115. Marks, A. E.; MOHAMMED-ALI, A.; PETERS, J.; ROBERTSON, R. (1985): The Prehistory of the Central Nile Valley as Seen from Its Eastern Hinterlands: Excavations at Shaqadud, Sudan. J. Field Archaeology 12, 261-278. NOoMINA ANATOMICA VETERINARIA, 3rd ed., NomIna HıstoLocıca, 2nd. ed. (1983): Published by the International Committee on Veterinary Gross Anatomical Nomenclature under the financial responsibility of the World Assocation of Veterinary Anatomists. Ithaca, N. Y. OBoussIER, H.; Ernst, D. (1977): Der formative Einfluß des Lebensraumes auf das postcraniale Skelett der Tragelaphinen (Tribus Tragelaphini Sokolov, 1953 — Bovidae, Mammalia). Zool. Jb. Syst. 104, 203-238. PETERS, J. (1986): Bijdrage tot de archeozoölogie van Soedan en Egypte. Ph. D. Diss., Rijksuniver- sıteit Gent. — (1986b): A revision of the faunal remains from two Central Sudanese sites: Khartoum Hospital and Esh Shaheinab. Archaeozoologia, Melanges publies ä l’occassion du 5° Congres international d‘Archeozoologie, Bordeaux-aoüt 1986, 11-35. — (1986c): Osteomorphology and osteometry of the appendicular skeleton of African Buffalo, Syncerus caffer (Sparrman, 1779) and cattle, Bos primigenius f. taurus Bojanus, 1827. Ghent: Occasional papers, Laboratorium voor Paleontologie, Rijksuniversiteit Gent, No. 1. The appendicular skeleton of African buffalo and of domestic cattle 123 STÖCKMANN, W. (1975): Die Form der Mandibel Afrikanischer Bovidae (Mammalıa) und ihre Beeinflussung durch die Ernährung. Ph. D. Diss., Universität Hamburg. VAN NEER, W. (1981): Archeozoölogische studie van Matupi (Ijzertijd en Late Steentijd) en Kiantapo (Ijzertijd) ın Zaire. Ph. D. Diss., Katholieke Universiteit Leuven. WALKER, R. (1985): A Guide to the post-cranıal bones of East African anımals. Norwich, England: Hylochoerus Press. Author’s address: Dr. Jorıs PETERS, Institut für Palaeoanatomıe, Domestikationsforschung und Geschichte der Tiermedizin, Schellingstraße 10/II, D-8000 München 40 Z. Säugetierkunde 53 (1988) 124-125 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 WISSENSCHAEIFELCHETKURZMELTERSUNNG EN Note on the skull size in the two sympatric Mouse Deer species, Tragulus javanicus (Osbeck, 1765) and Tragulus napu (F. Cuvier, 1822) By MADELEINE VAN DORT Institute of Taxonomic Zoology, Zoological Museum, University of Amsterdam, the Netherlands Receipt of Ms. 15. 6. 1987 At present, within the family Tragulidae, the sister-group of the Cervidae, four species are recognized: viz. Hyemoschus aquaticus (Ogılby, 1841), living in tropıcal West Africa and Central Africa north of the Congo River; Tragnlus meminna (Erxleben, 1777), occurring on Sri Lanka (Ceylon) and in the south of peninsular India, and the largely sympatric species Tragulus javanicus (Osbeck, 1765) and Tragulus napu (F. Cuvier, 1822), both found in southeast Asıa and on islands west of the Strait of Makassar. The last two species strongly resemble each other. The only clear differences are found in their colour pattern, mainly of the throat area, and in size; T. javanıcus being the smaller species and T. napu the larger one. Size differences are most distinct where the species live side by side, whereas ın allopatric populations, 7. javanıcus can even be larger than T. napu. When two species overlap geographically, the situation in which the differences between them are accentuated in the zone of sympatry, and weakened or lost entirely ın parts of their ranges outside this zone ıs called: Character displacement; see BROwn and Wırson 1956. T. javanıcus and T. napn have been suggested (Dr. A. C. V. van BEMMEL — in verbis) to be a nice example of this phenomenon, however without giving exact data. During the first part of a study of the taxonomy and speciation of Tragulidae, over 200 skins and skulls of T. javanıcus and T. napu have been studied, forming part of the collections of the Institute of Taxonomic Zoology (Zoological Museum) in Amsterdam, the Rijksmuseum van Natuurlijke Historie at Leiden and the British Museum (Natural History) in London. The author is grateful to the curators of the mentioned collections for the permission to study the material. The differences in the dimensions of the skins of the two species were apparent and often documented by the measurements provided by the collectors. Nevertheless, to avoid possible mistakes caused by different ways of preparing the skins and taking the measure- ments, only the skull dimensions (taken by the author) were used. To simplify the graphic representation of the results, only the condylobasal lengths are given (see figure 1). Comparing the condylobasal lengths of the skulls of allopatric populations with those of sympatric populations, it ıs obvious that character displacement in skull size is present in the two Mouse Deer species studied. It is necessary, however, to discuss also other influences which may have caused the observed differences. First the rule of BERGMANN, which assumes that larger body size ıs selectively advantageous in colder climates. According to figure 1, this could be a possible explanation for the differences found ın the more northern mainland populations and the more southern populations of Sumatra and Borneo of T. napu, were it not, that the differences in climate between the mentioned areas are slight. All the areas are within the region of tropical rain forests, although in the north U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5302-0124 3202500 Skull size in Tragulus javanıcus and Tragulus napu 125 H100°/o 75 Tragulus javanicus 50 Main land; Allopatric L25 n=12 0/100°/o Tragulus javanicus 75 50 and Tragulus napu Main land; Sympatric 25 n=20 0/100°%c Tragulus javanicus 75 and Tragulus napu 50 Sumatra/Borneo; Sympatric 25 n=100 0/100°/0 75 Tragulus javanicus 50 Java; Allopatric 25 n=26 0/100%0 Tragulus javanicus 75 and Tragulus napu 50 Small Islands; Sympatric 25 n=21 0/100%0 75 50 Tragulus javanicus and Tragulus napu 25 Small Islands; Allopatric n=4] 0 75 80 85 90 95 100 105 110 115 120 N 5 7 mm C.b.l. T. javanicus T.napu KA Both species Fıg. 1. Condylobasal lengths of 220 skulls of Tragulus javanicus and Tragulus napu, living allopartri- cally or sympatrically somewhat more influenced by the different monsoons than in the south. Another phenomenon to be taken into account is that in mammals, representatives of species occurring on small islands sometimes tend to be smaller than members of the same species living on large islands and on the mainland. This phenomenon may play a role in T. napu, but not in 7. javanicus (see figure 1). It can be ıgnored with regard to the differences observed between Tyagulus species living on small islands, sympatrically or alone. Summarizing, Tragulus javanıcus and Tragulus napu show character displacement in skull size, when living sympatrically. They tend to reach more intermediate sizes when living without congeners. Reference Brown, W.L., JR.; Wırson, E. ©. (1956): Character displacement. Syst. Zool. 5, 49-64. Author’s address: Drs MADELEINE VAN DORT, Dept. of Mammals, Institute of Taxonomic Zoology, Zoological Museum, University of Amsterdam, Mauritskade 61, NL-1092 AD Amsterdam, the Netherlands Z. Säugetierkunde 53 (1988) 126-127 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Deutsches Primatenzentrum, Göttingen Nachtaktivität von Spitzhörnchen (Tupaia belangeri) Von J. KurrE und E. Fuchs Eingang des Ms. 23. 9. 1987 Zur Familie der Tupaiidae gehören mit Ausnahme von Ptzlocercus lowii nur tagaktive Arten (LuckErTT 1980). Die Aktivität von Tupaia glis während der Hellphase erscheint u.a. darin begründet, daß ıhre Retina nur aus Zapfen besteht (SAMORAJsKY et al. 1966). Berichte über mögliche Nachtaktivitäten verschiedener Tupaia-Arten erscheinen teilweise widersprüchlich. SPRANKEL (1961) fand für Tupaia glis, daß Männchen nachts ohne Unterbrechung schlafen. Nachtaktivität beobachtete er nur bei hochträchtigen Weibchen; auch Inou£ und Honpa (1974) geben Nachtaktivität für weibliche Tupaia glis an. SORENSON und CoNAWAY (1964, 1966) fanden bei Direktbeobachtungen für T. longipes, T. gracıilis, T. chinensis und T. tana keine Nachtaktivität. Dagegen beschreiben Lıu et al. (1982), ohne Angaben zur Methode zu machen, für Tupaia belangeri (chinensis) geringe Nachtaktivität. Bei Untersuchungen individualtypischer Aktivitätsmuster von Tupaia belangeri stellten wir mit Passiv-Infrarotdetektoren (PID) fest, daß trotz völliger Dunkelheit ın den Hal- tungsräumen einige Tiere nachts aktiv waren. Beobachtet wurden insgesamt 16 adulte männliche Spitzhörnchen(Tupaia belangeri) aus dem Deutschen Primatenzentrum, Göttingen. Die Tiere lebten in vollklimatisierten Räumen einzeln in Edelstahlkäfigen, an deren Außenseite je ein hölzerner Nistkasten angebracht war. Zwischen benach- barten Käfigen befanden sich Sichtblenden. Die Lichtphase begann um 8.00 h und endete um 20.00 h. Futter (Tupaia-Standard-Diät, Altromın, Lage) und Wasser waren ad lıb. vorhanden. Die lokomotorische Aktivität wurde mit Passiv-Infrarot-Detektoren (PID) gemessen. Der Sensor dieses Systems (Superrot SR 2000 Passıv-Infrarot-Detektor, Visonic Ltd., Tel Aviv, Israel) bestand aus einem infrarotempfindlichen, pyroelektrischen Doppelelement, das gegen Luftströmungen und Anderungen der Umgebungstemperatur unempfindlich war. Es zeigte jedoch das Eindringen eines wärmeausstrahlenden Tieres in den vom Sensor abgetasteten Bereich an. Die Signale wurden als Summe in 10-Minuten-Intervallen von einem Epson Hand-Held HX 20-Computer registriert und auf einer Mikrokassette gespeichert. Von da wurden die Daten auf einen IBM-Personal Computer übertragen und mit „Lotus 1-2-3“ (1983 Lotus Development Corporation) graphisch bearbeitet. Einzelheiten zur Methode s. LERCHL (1986). Bei sechs von 16 untersuchten Tieren konnte mit der PID-Registrierung Nachtaktivität festgestellt werden. Die Tiere verliefen in unregelmäßigen Abständen bis zu achtmal pro Nacht für bis zu 20 min ihren Nistkasten. Ein Beispiel (8 855) ist in Abb. 1 dargestellt. Von zwei Tieren (8 855 und d 1008) wurden mit Video je vier Nächte aufgezeichnet und ihre Nachtaktivitäten quantitativ und qualitativ untersucht. Wegen der völligen Dunkelheit in den Haltungsräumen beleuchteten wir die Käfige mit Infrarot-Lampen (Filter > 840 nm; Göttingen Farbfilter GmbH) und benutzten zur Aufzeichnung eine Infrarot-Kamera (Grundig SAE 70). Während der Nacht wurden die Tiere weder durch die Aufnahmen noch den Kassettenwechsel am Videorekorder gestört. | Beide untersuchten Tiere (d 855; 8 1008) waren nach Ende der Lichtphase noch bıs zu 10 min aktıv (20.00-20.10 Uhr) bevor sie in ihre Nistkästen gingen. d 1008 verließ seinen Nistkasten während der vier Nächte insgesamt dreimal für zwei bis sechs min, d 855 insgesamt siebenmal für drei bis sechzehn min. An Bewegungsformen konnten nur Laufen und Klettern beobachtet werden. Das sonst für T. belangeri typische Springen trat ın der Nacht nicht auf. Insgesamt waren die Bewegungen langsam und vorsichtig. Beim Laufen U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5302-0126 $ 02.50/0 Nachtaktivität von Spitzhörnchen (Tupaia belanger:) 127 80 60 40 20 Impulse / 10 Minuten 0) (0) L 8 12 16 20 2L Licht aus Licht an Licht aus Tageszeit Motorische Aktivität von d 855, registriert über 24 h. Die Hellphase dauerte von 8.00-20.00 Uhr über Äste behielten die Tiere mit dem Bauch meist Kontakt zum Untergrund. Weiter wurden Fressen (max. 3:20 min), Trinken (max. sechsmal) und Körperpflege (max. 5:30 min) registriert. Vereinzelt konnten auch Markierverhalten sowie Koten beobachtet werden. Unsere Beobachtungen zeigen, daß innerhalb der tagaktiven Arten der Tupaiidae, zumindest für Tupaia belangeri, Nachtaktivität generell nicht ausgeschlossen werden kann. Vergleicht man unsere Ergebnisse mit denen von Liu et al. (1982) und anderer Autoren, so scheint die Erfassung und Beschreibung von Nachtaktivität von den eingesetz- ten Beobachtungs- und Registriertechniken abhängig zu sein. Danksagung Herrn Prof. Dr. ERKERT, Universität Tübingen, danken wir für die Überlassung der Infrarot-Kamera. Literatur Inou£, $S.; Honpa, K. (1974): A special Tupaia cage for the continuous detection of locomotor activity under the auto- or the self-control of light-darkness rhythms. Reports Inst. Medical & Dental Engineering 8, 145-147. LERCHL, A. (1986): Wirkungen exogener Melatoninchargen auf renale Steroidausscheidung und lokomotorische Aktivität bei Weibchen und Männchen von Sattelrückentamarinen (Saguinus fuscicollis HOFFMANNSEGG). Dipl.-Arbeit, Göttingen. Liu, $S.; WENYUAN, Z.; ZESHENG, W.; XıuJu, D. (1982): Studies on the circadıan rhythm of tree shrews (Tupaia belangeri chinensis). Sleep 1982, 6th Eur. Congr. Sleep Res., Zürich 1982, 226228. LuckETT, W.P. (ed) (1980): Comparative biology and evolutionary relationships of tree shrews. New York, London: Plenum Press. SAMORAJSKI, T.; ORDY, J. M.; KEEFE, J. R. (1966): Structural organization of the retina in the tree shrew (Tupaia glıs).]). Cell. Biol. 28, 499-504. SORENSON, M. W.; Conaway, C. H. (1964): Observations of tree shrews in captıvity. Sabah Soc. ]. 2, TlzaN: — — (1966): Observations on the social behavior of tree shrews in captıvity. Folia primat. 4, 124-145. SPRANKEL, H. (1961): Über Verhaltensweisen und Zucht von Tupaia glis (Diard 1820) in Gefangen- schaft. Z. wiss. Zoll. 165, 186-220. Anschrift der Verfasser: JoacHıMm KURRE und Dr. EBERHARD Fuchs, Deutsches Primatenzentrum, Kellnerweg 4, D-3400 Göttingen BIELEENBIESIPIRSE/EIFNEEN IESERN ZWEERS, G. A.; DULLEMEYER, P. (eds.): Architecture in living structures. Spec. publ. of Acta Biotheoretica Vol. 34. Boston, Lancester, Dordrecht: Nijhoff Publishers BV 1985. 184 pp. US $ 45.00. ISBN 90-247-3240-9 Die Veröffentlichung der 13 Beiträge zum 19. Lochmühle Symposion behandeln durchweg Fragen zur Methodik und Theorie einer Konstruktionsmorphologie (Funktionelle Morphologie i.S. der Leidener Arbeitsgruppe). Vorträge sehr verschiedenartig orientierter Forscher berichten über neue Ergebnisse unter funktionalen, strukturalistischen und transformistischen Aspekten und unter Heranziehen von Beispielen vorwiegend aus Forschungen an niederen Vertebraten (ein Beitrag handelt über Arthropo- den). Der Begriff der Konstruktionsmorphologie geht auf den bedeutenden Entomologen HERMANN WEBER zurück, dessen grundlegende Arbeiten bisher leider in der Morphologie der Wirbeltiere meist übersehen wurden. Es ıst dankbar zu begrüßen, daß durch den Beitrag von W. E. Reır, R. D. K. THomas und M. Fischer mit Nachdruck auf das Werk WEBER’s verwiesen wird, auch wenn dieses noch in die Zeit vor der allgemeinen Konzeption einer synthetischen Evolutionstheorie fällt und dadurch die Reserve WEBER’s gegenüber einseitigen Überlegungen zur Evolutionsfrage verständlich werden. Der genannte Beitrag ıst zugleich der einzige, der ein Beispiel aus der Morphologie der Säugetiere (Konstruktionsmorphologie und Stammesgeschichte der Hyracoidea durch M. FISCHER) einbezieht. Die Ausführungen von DULLEMEYER legen in klarer und weitgespannter Form die Problematik des Themas, die Methodik und wissenschaftstheoretische Einordnung dar. Gegenüber einigen stark reduktionistischen Auffassungen einiger Funktionalisten wird der integrative Charakter der modernen Konstruktionsmorphologie klargestellt und die Komplementarität verschiedener Betrachtungsweisen hervorgehoben. D. STARcK, Frankfurt/M. NACHTIGALL, W. (Hrsg.): Bat flight - Fledermausflug. BIONA report 5. Stuttgart, New York: Gustav Fischer Verlag 1986. XII, 235$. Zahlreiche Abb. u. Tab. DM 38,-. ISBN 3-437-20372-X, ISSN 0930-0635 Der jetzt vorliegende fünfte BIONA-Report faßt — nach drei Bänden über den Vogel- und den Insektenflug und einem Werk über Thermobiologie — die Ergebnisse eines Symposiums zusammen, das im Mai 1984 in Saarbrücken unter der Leitung von WERNER NACHTIGALL stattfand. Themen- schwerpunkt des Buches ist der Fledermausflug, doch sind in diesen Report auch Fragen der Flugbiophysik von Flugsauriern und Gleitbeutlern mit aufgenommen. Zusammen mit den bereits erschienenen Bänden 1-3 liegt damit jetzt eine Übersicht vor, die Probleme des Fluges aller wichtigen fliegenden Tiergruppen behandelt. Acht Kapitel dieses Buches, jeweils den Sachgebieten „Evolution“ und „Physiologie“ zugeordnet, behandeln den Fledermausflug. Die Themen reichen von der reinen Biophysik, über Funktionsmor- phologie und physiologische Energetik, bis hin zur mutmaßlichen Evolution des Chiropterenfluges. So wird die Evolution des Fluges (K. SCHOLEY) und die Entwicklung von Flug und Flügelformen (U. M. NORBERG) untersucht. Mechanik, Aerodynamik und Evolution des Fledermausfluges sind Thema eines weiteren Kapitels (]. RAYnEr). Die Untersuchungen des Energiebudgets von blütenbesuchenden Fledermäusen zeigen, wie unermüdlich diese kleinen Tiere auf Nahrungssuche sein müssen, um überleben zu können (O. v. HELVERsEn). Die Korrelation von Flügelschlag und Lautaussendung wird in zwei weiteren Arbeiten beschrieben (K. HEBLIcH, H. HERBERT). Vier Artikel, unter dem Stichwort „Gleiter“ zusammengefaßt, beschäftigen sich mit der Lokomo- tion des Taguan (Petaurista petanrista) (K. SCHOLEY) und einer allgemeinen Abhandlung über Flügelbaupläne (D. SCHALLER) sowie mit zwei typischen Bionik-Ihemen: So wird unter anderem anhand von Rauchkanalexperimenten aufgezeigt, daß Felloberflächen gegenüber glatten Flächen eine aerodynamisch günstige, grenzschichtstabilisierende Wirkung haben und die technische Nutzung dieser Erkenntnis diskutiert (W. NacHTIGaLı). Als Randgebiet des Themenkreises ist schließlich die Beschreibung eines vor fast hundert Jahren gebauten Fledermausflugzeuges mit in diesem Band enthalten (B. KresLing). Das Buch, das neben den genannten Arbeiten von VON HELVERSEN, NACHTIGALL und NORBERG, Beiträge einer Reihe jüngerer Autoren enthält, wendet sich an Leser, die mit den Grundproblemen des (Fledermaus-) Fluges vertraut und an weiteren Einzelheiten zu diesem Thema interessiert sind. EVELYN SCHLOSSER-STURM, Hamburg Deutsche Gesellschaft für Saugetierkunde 61. Hauptversammlung in Berlin, 27. September bis 1. Oktober 1987 Kurzfassungen der Vorträge und Posterdemonstrationen. Herausgegeben von Prof. Dr. Heinz-Georg Klös, Dr. Hans Frädrich, Prof. Dr. Carsten Niemitz, alle Berlin. Zusammenstellung: Christel Schmidt, Bonn. 1987. 57 Seiten. 24,5x 16,5 cm. Kartoniert 24,- DM Nach 30 Jahren fand dıe Hauptver- sammlung der Deutschen Gesellschaft für Säugetierkunde wieder in Berlin statt. Die überaus große Zahl der an- gemeldeten Vorträge und Poster be- wies, daß die nunmehr 750jährige Metropole - sie ist Gründungsort der Gesellschaft - für deren Mitglieder und Freunde an Anziehungskraft nichts eingebüßt hat. Säugetierfor- schung wird in Berlin an verschiede- nen Fachbereichen der Freien Univer- sität, der Technischen Universität sowie im Zoologischen Garten betrie- ben; ein Naturkundliches Museum, das nach der Teilung im Westteil der Stadt lange fehlte, befindet sich nun im Aufbau. Ein Hauptziel der Deutschen Gesell- schaft für Säugetierkunde ist, aufihren Jahrestagungen über Säugetiere arbei- tende Wissenschaftler verschiedenster Fachrichtungen zusammenzuführen, den Gedanken- und Erfahrungsaus- tausch anzuregen, um so Erkenntnisse aus den einzelnen Forschungsgebieten zu integrieren. Kommunikation, Fort- pflanzung und Funktionsmorphologie der Säugetiere waren 1987 die Schwer- punkte, über die berichtet wurde. Sie bieten durch ihre Vielfalt die Grund- lage für einen intensiven Gedanken- und Erfahrungsaustausch, eines der Hauptziele der Gesellschaft. Der Wunsch der Veranstalter, daß Berlin als Stätte des Dialoges den geeigneten Rahmen bilden würde, hat sich voll erfüllt. Zu beziehen durch jede Buchhandlun Verlagsbuchhandlung PA)L Paul Parey Spitalerstraße 12 DDr 2000 Hamburg 1 Erscheinungsweise und Bezugspreis 1988: 6 Hefte bilden einen Band. Jahresabonnement Inland: 308,- DM zuzüglich 13,80 DM Versandkosten, Gesamtpreis 321,80 DM einschließlich 7% Mehrwert- steuer. Jahresabonnement Ausland: 308,- DM zuzüglich 18,- DM Versandkosten. Das Abonne- ment wird zum Jahresanfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, Spitalerstraße 12, D-2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesellschaft für Säugetier- kunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. Z. Säugetierkunde 53 (1988) 2, 65-128 Das klassısche Werk über dıe Anatomie der Wirbeltiere Dieses anschauliche Werk ist vor allem für Studenten der Biologie, der Medizin und verwandter Studienrichtungen gedacht; es gibt aber auch den Spezialisten auf engeren Forschungsgebieten der Biologie zuverlässige Grundlagen für ihre wissenschaftliche Arbeit. Darüber hinaus wendet es sich an jeden, der am Beispiel der Wirbeltieranatomie die wundervolle Mannigfaltigkeit der Organismen zu verstehen sucht. %* Alfred Sherwood Romer/ Thomas S. Parsons: Vergleichende Anatomie der Wirbeltiere. Aus dem Amerikanischen übersetzt und bearbeitet von Hans Frick. Mit einem Geleitwort von Dietrich Starck. 5., neubearbeitete und-erweiterte Auflage. 1983. 639 Seiten mit 1310 Einzeldarstellungen, davon 123 farbig, in 435 Abbildungen. Gebunden 48,- DM Verlag Paul Parey Spitalerstraße 12 2000 Hamburg 1 PAUL P/REV Vol. 53 (3), 129-192, Juni 1988 ISSN 0044-3468 C21274F ZEITSCHRIFT FÜR SAUGETIERKUNDE INTERNATIONAL JOURNAL Or MAMMALIAN BIOLOGY eo 187 der Deutschen Gesellschaft für Säugetierkunde Mamm, Bolles, Kathryn: Evolution and variation of antipredator vocalisations of Antelope squirrels, Ammospermophilus (Rodentia: Sciuridae). — Zur Evolution und Variation der Warnrufe der nordamerikanischen Zieselgattung Ammospermophilus (Rodentia: Sciuridae) 129 Winking, H.; Dulic, Beatrica; Bulfield, G.: Robertsonian karyotype variation in the European house mouse, Mus musculus. Survey of present knowledge and new observations. — Karyotypvariation durch Robertsonsche Translokationschromosomen bei der europäischen Hausmaus, Mus musculus. Eine Übersicht über den derzeiti- gen Wissensstand und neue Informationen 148 Wirtz, P.; Kaiser, Petra: Sex differences and seasonal variation in habitat choice in a high density population of Waterbuck, Kobus ellipsiprymnus (Bovidae). — Geschlechtsunterschiede und jahreszeitliche Variation in der Habitatwahl in einer Hochdichte-Population des Wasserbocks (Kobus ellipsiprymnus) 162 Ferrari, C.; Rossi, G.; Cavani, C.: Summer food habits and quality of female, kid and subadult Apennine chamois, Rupicapra pyrenaica ornata Neumann, 1899 (Artiodactyla, Bovidae). — Beschaffenheit und Qualität der Sommer- nahrung von Weibchen, Kitzen und Subadulten der Apenningemse, Rupicapra pyrenaica ornata Neumann, 1899 (Artiodactyla, Bovidae) 170 Tiba, T.; Sato, M.; Hirano, T.; Kita, I.; Sugimura, M.; Suzuki, Y.: An annual rhythm in reproductive activities and sexual maturation in male Japanese serows (Capricornis crispus). — Jahresrhythmische Veränderungen von Fortpflanzungstätigkeiten und Geschlechtsreife beim männlichen Japanischen Serau (Capricornis crispus) 178 Wissenschaftliche Kurzmitteilung Ibanez, C.; Delibes, M.; Castroviejo, J., Martin, Rosalia; Belträn, J. F.; Moreno, S.: An unusual record of Hooded seal (Cystophora cristata) in SW Spain. - Ein ungs) swigialf] at Klappmütze (Cystophora cristata) in SW-Spanien 4 NN N 189 Buchbesprechungen 191 Verlag Paul Parey Hamburg und Berlin V HERAUSGEBER/EDITORS P. J. H. van BREE, Amsterdam -— W. FIEDLER, Wien — H. Frick, München — W. HERRE, Kiel - H.-G. Krös, Berlin - H.-]. Kunn, Göttingen - E. KuLzEr, Tübingen - B. Lanza, Florenz — J. NIETHAMMER, Bonn — H. REICHSTEIN, Kiel -— M. Rönrs, Hannover - D. STArcK, Frankfurt a M. - F. Strauss, Bern -— E. Tuenıus, Wien -— P. VoceL, Lausanne SCHRIETLEITUNG/EDITORTATZOFFECE H. SCHLIEMANN, Hamburg - D. Kruska, Kiel This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con- tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information Die Zeitschrift für Säugetierkunde veröffentlicht Originalarbeiten und wissenschaftliche Kurzmittei- lungen aus dem Gesamtgebiet der Säugetierkunde, Besprechungen der wichtigsten internationalen Literatur sowie die Bekanntmachungen der Deutschen Gesellschaft für Säugetierkunde. Verantwort- licher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Harald Schliemann. Zusätzlich erscheint einmal im Jahr ein Heft mit den Abstracts der Vorträge, die auf der jeweiligen Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde gehalten werden. Sie werden als Supplement dem betreffenden Jahrgang der Zeitschrift zugeordnet. Verantwortlich für ihren Inhalt sind ausschließlich die Autoren der Abstracts. Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z. Hd. Prof. Dr. Dieter Kruska, Institut für Haustierkunde, Biologie-Zentrum, Neue Universität, Olshausenstr. 40-60, D-2300 Kiel. Für die Publikation vorgesehene Manuskripte sollen gemäß den „Redaktionellen Richtlinien‘ abgefaßt werden. In ihnen finden sich weitere Hinweise zur Annahme von Manu- skripten, Bedingungen für die Veröffentlichung und die Drucklegung, ferner Richtlinien für die Abfassung eines Abstracts und eine Korrekturzeichentabelle. Die Richtlinien sind auf Anfrage bei der Schriftleitung und dem Verlag erhältlich. Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der Korrekturfahnen erfolgen. Vorbehalt aller Rechte: Die ın dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich | geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei- ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grund- sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch ım Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber- rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sınd mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of the fırst page of an article in this journal indicates the copyright owner’s consent that copies of the article may be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective, or for resale. For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the GE: Mit zwei Beilagen des Verlages Paul Parey Fortsetzung 3. Umschlagseite © 1988-Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstr. 12, D-2000 Hamburg 1; „Lindenstr. 44-47, D-1000 Berlin 61. — Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., : D-2240 Heide/Holst. Z. Säugetierkunde 53 (1988) 129-147 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Evolution and variation of antipredator vocalisations of Antelope squirrels, Ammospermophilus (Rodentia: Sciuridae) By KATHryNn BOLLES Department of Biology, University of Calıfornia, Los Angeles, USA Receipt of Ms. 12. 01. 1987 Abstract Analysis of variation in antipredator (alarm) vocalisations of North American antelope squirrels (Ammospermophilus) indicates the existence of significant differences among species, with little variation present among individuals, sexes, seasons and ambient temperature within species. Stepwise discriminant analysis of ten sonograph measurements for five species samples indicates that mean pulse duration, frequency of main sound energy, and frequency at beginning contribute most to species separation. The short harsh trill of A. interpres, A. nelsoni, and A. insularis appear to be adapted to closed habitats (rocky/prairie), and the long pure-toned trill of A. harrisii and A. leucurus to open desert conditions. Comparisons with the vocalısations of other sciurids suggest that the short calls are primitive, and that the evolution of long trills has accompanied the recent development of deserts in North America. Introduction There is widespread interest in the importance of vocalisations in the ecology and lıife- history of various groups of mammals (e.g., Kos 1981; RıEGER and PETERS 1981; SCHMIDT et al. 1982; PETERS 1982, 1984; HERBERT 1983; SIEBER 1985). Within the family Sciuridae vocalisations may constitute important factors in the evolution of sociality as they form the “glue” of their societies. Many descriptions and analyses of squirrel vocalisations examine aspects of varıation within populations, and individual response to differing stimulus situations (see OwınGs and HEnNEssy 1984) but fewer attempt to assess varıation among populations. Before the role of antipredator vocalısations (alarm calls) in sciurid ecology and evolution can be fully assessed it ıs desirable to understand more about the nature and levels of variation that exist within and among populations and species. This paper quantitatively analyses varıation in antipredator vocalisations wıthin and among the five living species of the genus Ammospermophilus, North American antelope squirrels, which have not been studied previously, and discusses the varıation from ecological and evolutionary viewpoints. Materials and methods This study is based on vocalisation data from a total of 57 adult individuals representing the five Ammospermophilus species, A. harrisıi (Har), A. leucurus (Leu), A. insularıs (Ins), A. interpres (Int), A. nelsoni (Nel), and Spermophilus variegatus and Sciurus niger (see specimens examined). In order to ensure that the samples measured included only adults, a survey was made to determine the condition of teeth and sutures in the skulls of individuals considered to be adults on the basis of their reproductive state (gravid or well-developed testes descended). All data are from adults by the criteria developed: fronto-parietal suture ankylosed (see also CALLAHAN and Davıs 1977) and the bregma completely closed (Harı 1926), yellow enamel on incisors, permanent upper P°-M° present in occluding position (HOFFMEISTER and DIERSING 1978; RoBINSON and HOFFMANN 1975) except in A. insularis, which may lack P? and its alveolus. In addition, all skulls exhibited tooth wear (BEG and HOFFMANN 1977; CALLAHAN and Davis 1977). All specimens were live-trapped and the vocalisations given in response to my presence immedi- ately after capture were recorded on a UUHER 4000L tape recorder (range 40-20000 = a U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5303-0129, $ kus U 4 1988 LIBRARIES 130 Kathryn Bolles per s (7.5 ips), using a unidirectional microphone, with a level frequency response of 50-12000 cps, held approximately 40 cm from the subject. A total of 155 sonagrams were made in the frequency ranges 80-8000 Hz and 160-16000 Hz from the recordings with a Kay Sonograph 6061B using wide- and narrow-band filters, to emphasise time and frequency respectively. The frequency spectra of the calls is shown as a function of time; amplitude is proportional to blackness of the impression. Time was measured with the wide filter bandwidth, and frequency with narrow filter bandwidth. Fifteen measurements were taken from the sonagraphs. Three highly correlated pairs of measure- ments were “redundant,” and one of each was selected for further analysis. Two additional characters proved to be highly variable and also were eliminated. The following measurements taken from the sonagraphs were used in the analyses (DOC, DOT, MPD in s, the rest in kHz): DOC - duration of call - length of call from first sound emitted to end of call; DOT - duration of trill - length of trill from first clearly configured pulse to end of call; MPD - mean pulse duration - the mean duration from the beginning of one pulse to the beginning of the next, measured at the beginning (from the third pulse), middle, and end of call; FRO - frequency band around trill at onset of call, the range of frequencies emitted simultaneously in” the noise characteristic of “kuk” (HorwıcH 1972); FMT - frequency of main sound energy of the trıll half way in its duration; FBT - frequency of trill at beginning; LFR - lowest frequency emitted; FRF - frequency of the fundamental; FRH - frequency of first harmonic above main sound energy; CAR - cascade ratio, the ratio of the pitch at the beginning of the call to the pitch at one-half the length, where the amplitude is high. Because some vertebrates exhibit variation in their vocalisations emitted at different temperatures, measurements were made to determine if vocal traits have any relationship to ambient temperature. The measured sonagrams for A. nelsoni and A. leucurus are from recordings made in the lab at 20°C and 31°C, with an hour difference ın time. Data for A. interpres and A. insularis are from samples recorded at two ambient temperatures for each population in the field. A. harrisii was examined for variation at two different times of the year, during the reproductive season, November-May (descended testes, gravidity, lactation) and from the non-reproductive season, June-October. Statistical analyses were carrıed out by use of BMDP programs (Dixon 1981). BMDP3D was used for univarıate analyses including the comparisons of the varıance of ten multiple calls of a single individual in each species with its respective species sample (50 F-tests) and for pair-wise comparisons among the five species (100 t-tests). These tests were applied also between all individuals of each sex in each species to estimate this possible source of varıation. Stepwise discriminant function analysis was performed with BMDP7M, F-to-enter set at 4.0. For each of the first four vectors (variates) a standardised canonical coefficient, indicating the relative weight of each character for each varıiate, was computed by multiplying the canonical coefficient for the character by its pooled standard deviation. _ Cluster analyses were performed with BMDP2M, using Euclidean distance with unweighted variables and by cladıstic analysıs. All specimens were prepared as standard skins, skulls, and postcranial skeletons to identify adults (of which vocalisations were measured), for future verification of species identities, and for adjunct studies of genetics and morphology, and ontogeny of vocalisations, and are deposited in the Natural History Museum of Los Angeles County: Ammospermophilus leucurus, 6 females, 4 males; A. harrisıı, 13 females, 5 males; A. nelsonz, 11 females, 11 males; A. interpres, 3 females; A. insularıs, two females, two males; Spermophilus variegatus, one male; Sciurus niger, one male. The small sample of four individuals of A. insularis, from a remote island, affects statistical analyses for similarıty between species using the t-test, with greater likelihood of supporting the hypothesis of no difference. Also, the statistics methods take into account varying sample sizes. Results The alarm call vocalisation given by antelope squirrels (Fig. 1) ıs a trıll as defined by BROUGHTON (1963), a rapid succession of repeated sounds or “syllables,” resolvable by the human ear, and has a narrow frequency envelope. The call may begin with a wide- frequency band of noise whose onset is sharp, and is followed immediately without a break in time by the more pure-toned trill. This initial rapıd burst of sound, measured as the frequency around the trill at onset (FRO), is referred to here as “kuk,” as it is sımilar in the structure of its wide range of frequencies to the “kuk” named and described by HorwıcH (1972) tor Sciurus carolinensıs. Ammospermophilus antipredator vocalisations 154 A. harrisii N NEN N I ETNNTEEPTAFERROTEORNURTEINELEUEIEER DE a we Ein. ni Il — DIET LIION TO Tee A. leucurus AHER I Na I rare EDER EI HEN EN) A. insularis KILO ER ITZ —-— DD W 2 OD N @ — D Wa Pa od N © Il Be In A. nelsoni - DD a 2 IN NO (0) 0.5 1.O 1.5 20 2.3 SECONDS Fig. 1. Sonagrams of representative antipredator vocalisation of each of the five species of Ammosper- mophilus Interspecific variation Sonagraphs of calls from the five species of Ammospermophilus (Fig. 1) suggest that there are differences with respect to frequencies of main sound energy and fundamentals, harmonics and noise, length of call, and distance (time) between pulses of the trill. Also, the main sound energy may be the fundamental, or it may be an harmonic of the fundamental; there may be one or more harmonics above the main sound energy. A. leucurus and A. harrisii have calls that appear long, and relatively pure-toned, without a wide band of frequencies emitted at the beginning. A. nelsoni, A. interpres, and A. insularıs appear to have a shorter call, with a wıde band ot frequencies at ıts onset. In addition, A. 132 Kathryn Bolles interpres appears to have a longer pulse duration than any other species, A. nelsoni has a shorter pulse duration than any other species, and both have a noticeably deeper-pitched trill. Univariate analysıs of the data for ten alarm call measurements (Tab. 1) indicates significant differences (.05 confidence level) among the calls of the species (Tab. 2) ranging from one character (between A. insularıs and A. interpres) to all ten (between A. harrisiüi and A. nelson:). All the call characters differ between at least one species pair. Duration of calls (DOC or mean DOC) is 2.13 s and 2.24 s in A. leucurus and A. harrisii respectively, and is approximately twice the length of the calls of A. nelsoni (1.05 s) and A. ınterpres (1.07 s), whereas A. insularis is intermediate (1.35 s). Duration of trill (DOT) ıs related to duration of call and varies from it by the duration of the kuk sound, if present. The kuk as measured by band of frequency at the beginning of the calls (FRO) in A. leucurus and A. harrisii ıs small (2.48 and 2.22 kHz), but in the other species is broad, producing a sudden, loud harsh noise at the beginning of the call. Frequency of the main sound energy of the trill (FMT), frequency of the first harmonic (FRH), and frequency at the beginning of the trill (FBT), are shared (1.e. not significantly different at .05 level) by A. leucurus, A. harrisii and A. insnlaris, the species with relatively high-pitched calls. These same three characters are shared by the two species with relatively low-pitched calls, A nelsoni and A. interpres. The lowest frequency emitted (LFR) and frequency of the fundamental (FRF) are shared among all the species, except A. nelsoni. The mean pulse duration (MPD) is shared by A. leucurus and A. harrisii (MPD = .052 and .050 s) and by A. harrısıl and A. insularis (MPD = .044). In A. interpes this interval is longer (MPD = .063) and ın A. nelsoni, half this length (MPD = .032). Table 1. Measurements of characters of alarm calls of 57 individuals of the five species of Ammospermophilus Character A. nelsoni A interpres A. insularıs A. leucurus A. harrısü Acronym N=22 N=3 N=4 N=10 N=18 DOC 1.052:=50099 1.070=270.08 1952084 2A38=-=R 9316 a2 Vslld 09002534 0.920120 0.8200 228 140723204 130223588 FRO 6.995 -=20.34 6.43: =2°0.47. 73322017 2.48 ==20:53 2222-0098 1.007227.89 3906277.00 7.007 7.80 0,205=520 0.605 20.50 MPD .032 .001 .063 .003 .044 .001 .052 .002 050 22001 .022 .038 .060 .070 .040 .048 .047 .062 050752057 FMT 1:.992719:05 1.997=220.56 2.39, 2500.45 2.5302 0414 227,3, 2280310 0.205=21r.80 1,20723.00 1.205339 1.5053209 1.805320 EAR 1507250:04 1.07222097 00) 2ER 1r192==29:08 1.OI=EI9.03 0.9221760 MOO—IE2O O99E E02 00016877 OrS Om BER 0.07 & 0.04 0.57 = .030 0.152==20.09 1212-3026 1-402--30518 0:.00520.50 0.00 — 1.00 0.00520.30 0.002250 0.007—E2760 FRF 1.05,:270:05 1192220719 1.002=270.35 168352021 2.032 2:2 08107 0.60 - 1.60 0.907 1.50 05102179 1.105220 1.107359 FBT 1:332==20.05 22002=:20:55 2,0 ae a 2.89,=220.07 2.69,==90203 1.007297, 1.2053:099 1.20003722 2.601 23,20 1.805330 FRH 27.08-220:15 9-10==90585 4:5 08-1175 1 as | 402220922 160525330 2.207 24:80 2230) 2110.39 3.7 025:30 3.49.7370 DOT 0.96 + 0.08 1.072:=20:08 1398==. 0,34 2ASS==30316 2DI,-=BOAN6 0,572 21260 0.925129 058202228 140523304 1.30553283 Means + standard errors shown over ranges. Character acronyms are explained in text; DOC, MPD, and DOT are time in seconds, others are frequencies in kilohertz Table 2. Matrix of alarm call characters in which Ammospermophilus species differ significantly Ammospermophilus antipredator vocalisations (P=.05), above diagonal, and those which they share (P > .05), below diagonal A. nelsoni A. interpres MPD, CAR, PERSEBIE FMT DOCHCHR JBEIRSHEIRO, ERTSEBIE A. insularis MDP, CAR, EB EMI, FRH A. leucurus DO@, MDB, IBERSSERO®, ERESEBTE EIMASSEREI, DOT DOC-MED) EROFEB/IE DOT DOCZMED, EFR,ERO, EB DO A. harrısü DOC, MPD, EARSEERS EROFTERE} EB ENGE FRH, DOT DOC, MDP, EROJEREN FBI5 EMI, FRH, DOT DOCIEER EROJERE, DOT FMT, FRH, DOT EARSTEER FRF, FMT, FRH CAR, LER CAR, FRF, EMISWERHI MPDEEARS DOCZMED) EBISSMET, EERTERO, FRH EREIEBIE FMT, FRH, DOT Character acronyms explained in text The cascade ratio (CAR) is the change in pitch from the beginning to the midpoint of the trill, where amplitude is high. These calls have a soft quiet end. This ratio of FBT/FMT is nearly 1.00 ın calls that do not have a drop in pitch (A. insularis, A. harrısü, and A. interpres) and higher of course in calls that cascade: A. nelsoni has a mean cascade ratio of 1.30, and A. leucurus, CAR = 1.19. To summarise, (Tab. 2) the alarm call characters that show the largest number of differences among the species of Ammospermophilus are mean pulse duration (MPD) (eight differences), frequency at beginning of trıll (FBT) (seven differences), frequency band at onset (FRO), duration of call (DOC), and trıll (DOT) (all with six differences). In duration of call (DOC) and frequency band at onset (FRO) A. leucurus and A. harrisu are similar, and distinguished from A. nelsoni, A. interpres, and A. insularıs in which these characters are similar. In mean pulse duration (MPD) and frequency of main sound energy (FMT) A. leucurus, A. harrısü and A. insularis are the same, and differ from A. nelsoni and A. interpres. Stepwise discriminant analysis of the ten sonagraph measurements for the five species of Ammospermophilus indicate significant separation between all five species centroids (Fig. 2). Eighty-nine percent of the total varıation is explained by the first two canonical variates. Standardised coefficients for the first four canonical varıates (Tab. 3) indicate that mean pulse duration (MPD) and frequency at beginning of the call (FBT)) make the greatest contribution to the first canonıcal varıate, MPD and frequency of the main sound energy, or pitch (FMT) to the second. Characters which never entered into the discriminant analysıs were DOC, DOT, FRF, and LFR. FRH is not used in the analysıs because it is not always present. The first canonical variate separates A. nelsoni from all other samples, primarıly by MPD and FBT. A. interpres ıs distinguished from all other species by the second canonical 134 Kathryn Bolles variate, with MPD and FMT making the largest contribu- tion. The distributions of in- dividuals are separated for all species, except for A. harrısıı and A. leucurus. The centroids for A. lenucurus and A. harrisu, however, are significantly different (D’= 16.16, P= .005). Ihe Kiseparanıonggon species by discriminant func- tion analysıs was mainly by mean pulse duration (MPD), and pitch (FMT and FBT). An a posterior classifica- tion (Tab. 4) correctly ıden- tified 90 % of the individuals. Three A. leucurus and two A. Canonical Variable | harrisii were cross-ıdentified Fıg. 2. Plot of the first two canonical varıates, based on antipre- and Er A. harrısü call va dator vocalisations, separating the five species of Ammosper- misassigned as 4. nelsoni. All mophilus. Star indicates population centroids. The first variate of the A. nelsoni, A. insularıs, (abscissa) represents 64 percent of the observed variance; the and A. interpres were correct- second (ordinate), 25 percent ly assıgned. The percentage of Variable 2 Canonical Table 3. Standardized coefficients for canonical variates for the first four vectors for alarm calls of 57 individuals of five species of Ammospermophilus Canonical varıates Character MPD FRO FMT FBT CAR % Variance explained Table 4. Classification matrix based on stepwise discriminant analysis of alarm calls of 57 individuals of the species of Ammospermophilus Species Percent Number of cases classified into group Correct LEU HAR INS . nelsoni 100.0 22 . leucurus 70.0 0 . harrisii 83.3 1 . insularis 100.0 0 ) . interpres 100.0 Total 89.5 23 Ammospermophilus antipredator vocalısations 155 correct classification indicates the high degree NEL of species specificity of alarm calls in Ammo- INT spermophilus. The alarm vocalısations of the genus fall INS into two general types: harsh (wide band of LEU frequencies at beginning), short, and with rela- tively low-pitched trill in A. nelsoni, HAR A. interpres and A. insularis; and longer rela- 5 a 2 2 | tively pure-toned and higher-pitched trills in een Call A. leucurus and A. harrısı. The greatest JK > number of similarities (Table 2) are between Fig. 3. Phenogram based on similarities of A. insularıs and A. interpres, and between A. antipredator vocalısation characters in the leucurus and A. harrisü. A. nelsoni has fıve five species of Ammospermophilus similarities each with A. interpres and A. in- sularis. A. nelsoni shares only one character with A. leucurus. A. nelsoni and A. harrısu differ in all ten characters. A. leucurus and A. harrisii, then, are similar, and A. nelsoni, A. insularis, and A. interpres have similar vocalisations. Cluster analysıs (Fig. 3) based on Euclidean distance (d;) between the five distinct population suggests close phenetic relationship between A. harrısu and A. lencurus, wıth A. nelsoni, A. interpres, and A. insularıs forming a second cluster. Intraspecific and individual variation For each of the five species, the varıance for ten calls of an individual of each species was compared with the respective species sample. In each instance the within-individual varıance (F-statistic, P <.05) is significantly lower. This suggests that the species popula- tions are separate taxa (species or subspecies). There appears to be no sexual dıvocalism (sensu MARSHALL and MARSHALL 1976) in the structure of calls of Ammospermophilus. There are no significant differences (P =.05) found between females and males in any character. Individual Ammospermophilus do not show any differences (P =.05) ın four call characters examined (Tab. 5, DOC, MPD, FMT, FRO) at different ambient temperatures. No significant differences were found for the four characters between individuals of A. harrısü collected in the reproductive season and those from the non-reproductive season (abs5): Table 5. Comparisons of means of characters of Ammospermophilus species calls recorded at (A) differing temperatures, and (B) differing times of year A B A. nelsoni A. leucurus A.interpres A. insularıs A. harrısu (KB 217) (KB 232) (KB 167) (KB 119) n=9 n=9 n=1 n=3 n=3 n=3 n=3 n=3 n>=3 n>=3 Nov.- June- 2IRSESIEEERIEC HZ IE ICH ESHPI SS BEE May October Pitch (FMT) 150504181550 22759, 273051 011 Sl 1NSEEE131°6 DNS Pulse rate (MPD) (03502..0547 27 0525:0511552.002252. 061 52:.039555.033 0517 57:048 Benschgorzealı DOC) 1.45 1.21 1.572 1.572 1.01 0.93 1.117 1.38 RAY 202 “Kuk” frequency 67.902 0.727.423 37437 2%9378.6:80 7750, 87:10 212,53 span (FRO) Temperatures are in degrees Celsius; seasons of year explained in text; FMT, FRO in kilohertz, MDP, DOC in seconds. KB are field catalog numbers 136 Kathryn Bolles Discussion Interspecific variation The vocalisations were recorded under similar circumstances and fulfill the criteria for homology of behaviors of TEMBROcK (1968): 1. the criterion of position (stimulus), 2. the criterion of special quality of structure (erill) and 3. the criterion of interconnection by intermediate forms (see transformation series discussed below). While an individual’s antipredator call may exhibit large variation concomitant with varying stimuli (are not structurally unitary) (OwınGs and HENNEssY 1984) and dialects are described in squirrel subspecies (SLOBODCHIKOFF and CoAsT 1980), and these sources of variation may occur in Ammospermophilus, it is important that the conditions of collecting the sound recording be controlled for the kind of comparisons made in this study and which yield small intrapopulation variation. Thus these alarm calls of the five species of Ammospermophilus are homologues and possibly differ due to genetic differences between the species populations. Although much work on birds suggest there ıs a large learning component in the forms of vocalisations (NOTTEBOHM 1972; KroopsMA 1982; KroopsMA and Bayııs 1982; Barprtısta and PETRINOVICH 1984) a genetic contribution may be larger in mammals. The most direct evidence for the inheritance of vocalisations in sciurids are the studies of interspecific hybrids in Tamiasciurus and Spermophilus that emit calls intermediate between those of the two parent species. Hybrid Tamiasciurus hudsonicus X douglasır emit three kinds of vocalısation, each intermediate in structure between those of the parents (SMITH 1978). Spermophilus richardsonu X elegans hybrids produce two types of vocalısa- tions, the characters of which were consistent for each individual. The call characters of some hybrids are intermediate between parental species calls, and in others resemble one of the parent species (KoeppL et al. 1978). Ecological function of alarm calls One source of the species specificity observed may be related to the functions of calls in different habitats. An important characteristic of alarm calls is the degree to which they may be located by the hearer. This function of a call may be achieved by its form (MARLER 1956; Lewis and Cores 1983; Lewis 1983). Broad band sıgnals, whose sonagrams appear vertical, short, with much noise and subharmonics and a wide range of simultaneously emitted frequencies, tend to be easily located. The calls of A. nelsoni, A. interpres, and A. insularis are of this form, and both conspecifics and predators might determine more easily the location of the caller, and the predator. Calls that result in pure-tone horizontal sonagrams are diffuse, and may be difficult to locate; this form is characteristic of calls of A. leucurus and A. harrisüu. Thus the call types of Ammospermophilus may have important ecological consequences. These two call types are associated with habitat types (Fig. 4). The association between physical aspects of the habitat and the structure of vocalisations is often observed (MARTIN and MARLER 1977; Wırey and RıcHarps 1978; PETERS 1984). The species that give the short, low-pitched call with broad frequency band at the beginning, A. nelsoni, A. interpres, and A. insularis, generally occur in more “closed” (non-open) habitats, rocky areas with dense vegetation (Fig. 5a-d), which occlude the view of a small ground squirrel. (They could also be interpreted as concealing habitats, or habitats with safe outlook perches.) A. leucurus and A. harrisii have a pure-toned call and occur in “open” sparsely vegetated xeric habitat (Fig. 5e, f). Interestingly, the association between habiıtat type and the length of call in the five Ammospermophilus populations contrasts with this trait and habitat in three populations ot Gunnison’s prairie dogs, Cynomys gunnisoni (SLOBOD- CHIKOFF and CoasTt 1980). Ammospermophilus antipredator vocalısations 137 OL cr Mlihbuhlien. mn j 4 , ONNDROODORODNOGAODTOnEn ohnnnonn BL 3 hi ‚| WTA, 8 > ER 3. A. insularis RL DD & -- 9 h Dana, Fıg. 4. Ranges in western United States and northern Mexico of the five extant species of Ammosper- mophilus with location of populations (X) from which vocalisations (insert) were analysed The habitat of A. nelson:, today and ın the past (]. SAUER, pers. comm.) is primarily grassland, often tall enough to hide a small anımal standing on the ground. Clearly, this habıtat is, for the squirrel, a “closed” (non-open) habitat (Fig. 5a, b). A. insularis occurs in highly rocky habitat with dense sub-tropical vegetation (Fig. 5d). It is abundant on rocky ridges and steep-walled canyons on Isla Espiritu Santo. The sarcocaulescent desert on the island approaches thornscrub ın density, and contains many trees (Cercidium, Bursera, Jatropha, Pachycereus) (WıGcıns 1980). Kathryn Bolles 138 ich the samples were collected for the five species of Ammosper- ıes from wh itats at localıt A and B . nelson Hab IKT Da A. leucurus A A. harrisii b) A. insularis D= interpres; = A 2; =A mophilus Ammospermophilus antipredator vocalısations 139 A. interpres occurs ın the Chihuahuan desert region, restricted to rocky foothills where it may be seen on tops of boulders, or in junipers. The species is generally absent from plains, plateaus and level, sandy terrain (FinpLey et al. 1975) and prefers the rough, dissected cliffs and rocks at 1100-2000 meters elevation, often with woodland vegetation (ScHmidry 1977). The population studied was found at 1800 meters elevation on rocky mountain slopes covered by grassy, pınon-juniper woodland (Fig. 5c), anon-open habitat where individual squirrels are not readily visible. The latter three species have a call with characteristics that suggest it is easily located. Possibly the calls relay awareness of the position of the predator to both conspecifics and the predator, while the squirrels have the relative security of concealment. A. harrisii and A. leucurus inhabit the Sonoran, Mojave, and Great Basın Deserts. While A. harrısii occurs extensively in rocky areas with saguaro and paloverde vegetation, populations often live on sandy flats with relatively sparse treeless vegetation (Larrea, Opuntia, acaulescent yuccas) such as the habitat from which the study sample was collected (Fig. 5e). Similarly, although A. leucurus inhabits a number of relatively closed habitats such as juniper-joshua tree woodland of the Mojave Desert, the species ranges extensively across open sandy treeless flats of Artemesia, Larrea or Atriplex ın the Sonoran and Mojave Deserts, where the study sample was collected (Fig. 5f). An individual A. leucurus or A. harrısü is usually visible when calling in a picket pin posture (GANDER 1967) from the ground, or from the top of a creosote bush (VAUGHAN 1954). To escape to a burrow it often must cover a wide expanse of open ground. The “tendency to run, rather than climb or seek nearby shelter, is reflected in the relatively longer feet of this species /A. lencurus) ... in contrast to... A. interpres, which spends very little time away from rocks” (FinDLey et al. 1975). Thus A. leucurus and A. harrısıı sound an alarm that gives minimal information about their spacial position, while the three non-open habitat species can afford having a more informative, easıly located call, which relays both position and alarm. In addition to the above determinants of frequency band width (FRO) and call length (DOC), another trait concomitant with habitat is pitch of trıll (FMT) for optimal transmission of sound. For maximum range of transmission, sounds produced within a few meters of the ground should be between one and four kHz in any habıtat (WırEy and RicHARDS 1978). These conclusions may explain the dominant mean frequency range of 1.09 to 2.73 kHz in alarm calls of Ammospermophilus, and suggest that transmission of their vocalisations may be optimal when they call from tops of large vegetation. While the differences ın habitats for species of Ammospermophilus are considerably less than between tropical forest and grassland, trıll pitch for four of the five species is consistent with the prediction of the MARTEN and MARLER (1977) model. A. leucurus and A. harrisii sampled from open desert have higher-pitched (2.50-2.73 kHz) calls. A. nelsoni, and A. interpres sampled from “non-open” habitat have calls with lower pitch (1.09-1.90 kHz). A. insularıs, also from non-open habitat, does not fit this pattern for frequencies of main sound energy, with a mean pitch of 2.33 kHz. Morphologic transformation A transformation series ıs presented (Fig. 6) that hypothesises the sequence of evolutionary changes that may link the calls of Ammospermophilus. The calls of A. interpres (Fig. 6d, e, f, g) show greater variation in most characters of their alarm calls (similar to the structurally graded series of LEGER and Owıncs (1978) and Owıncs and Hennessy (1984) than do other species of the genus. Individual calls of A. interpres are intermediate in form between calls of A. harrisüu and A. leucnrus which are at one end of the series, and the usual forms of A. interpres, A. insularıs and A. nelsoni at the other end. 140 Kathryn Bolles IL AOL UGS STREBEN: \MMERRERRPERPERRRERE KÄamn DT RRRRRRRRRREE ER ! UREUEUNWUHTEEUUURIRN None - DW aD .N © 5: nern an , | > ann Ipunı N N Me: 10 SECONDS Fıg. 6. Hypothetical transformation series for antipredator vocalisation structure in the five species of Ammospermophilus. Sonagrams are from a = A. harrisü; b and c = A. leucurus; d, e, f, and g = A. interpres; h and ı = A. insularıs; | = A. nelsoni Sonagrams of calls of A. interpres that illustrate such intermediate variation form the basis of the transformation. From this morphoseries the sequence of evolutionary changes, but not its direction, can be inferred. The existence of such a series is also one of the criteria for homology (TEMBROcK 1968). The calls of A. harrısi and A. leucurus (Fig. 6a, b, c) are high-pitched (FMT), long Ammospermophilus antipredator vocalısations 141 (DOC), with wide pulse rate (MPD) and no “kuk” (FRO). At the other end of the spectrum are calls of A. interpres (Fig. 68), A. insunlarıs (Fig. 6h), and A. nelsoni (Fig. 6)) that are short with a broad band of frequencies at the onset, the “kuk” sound. One trend between the extremes is a shift in frequency of the trill by a change in emphasis to a fundamental (Fig. 6g and j) or an harmonic (Fig. 6g and j) emerging as fundamental (Fig. 6a and b). In general, there is gradation from the short pulse rate in short harsh calls (Fig. 6h and j) to the longer pulse rate of long pure-tone calls (Fig. 6a and b). Although the gross appearance of the A. insularıs call ıs similar to that of A. interpres and A. nelsoni, its pulse rate is longer (Fig. 6h). The pulse rate of A. insnlaris ıs not statistically different (P =.05) from those of A. leucurus and A. harrısii (Fig. 6a, b, c). The pitch also does not change (statistically) in this morphoseries from A. insularıs to A. leucurus-A. harrisii. The range of variation in these characters suggests that A. insularıs, A. leucurus, and A. harrisii are not different in MPD. Also, A. insularis, and A. interpres may be transitional between A. nelsoni and A. lencurus-A. harrısu. Phylogeny based on antipredator vocalisations The utility of behavior as a taxonomic tool for the establishment of phylogenetic relationships among rodents has been largely neglected, although species-specific calls have been described for a number of groups (e.g. CLARK 1974; GREENE 1978; GREENE and BURGHARDT 1978; KoeppL et al. 1978; MAaTocHA 1975; Nıkouskıı 1974, 1976, 1979; SMITH 1978), and evolutionary interpretations based on other behaviors have appeared for other vertebrates (e.g., CoLLıas and CorLıas 1977; MARSHALL and MARSHALL 1976; McCarLey and McCaAR- LEY 1976; STRUHSAKER 1970). To utilize the call trans- formation series to esti- mate phylogenetic rela- tionships and derived characters it ıs rooted by outgroup comparison. Al- though many specific be- haviors could be compared for this purpose, vocalısa- tions are ideal because they may be visually and quantitatively represented on the sound spectrogram. Also, alarm calls are one of the communication be- haviors of squirrels which are important in their so- cial structure and evolu- tion (SHERMAN 1977). Trills similar to those of Ammospermophilus in ı occur in a few other sci- Rh ha urids. Those which resem- 0 05 1O 1.5 20 ble them are often pre- SECONDS ceded by a sound, whose Fig. 7. Sonagrams of antipredator vocalisations of a = Spermophilus onset ıs sudden and com- variegatus, b = Sciurus niger, c = Ammospermophilus nelsoni Ze DEU EDENEONTENE® KULOAERITZ a N N © —- D Wa +» li - DW. a 9 N © nr 142 Kathryn Bolles posed of a wide band of simultaneously emitted frequencies. These are emitted by chipmunks, Eutamias (BRAND 1976; CALLAHAN 1980; DUNFORD and Davis 1975), grey squirrels, Sciurus carolınensiıs (HorwıcH 1972) and fox squirrels, Sciurus niger (ZELLEY 1971). I recorded similar vocalisations (Fig. 7) given by Sciurus niger (fox squirrel) and Spermophilus (Otospermophilus) variegatus (a rock squirrel) that appear to be homologous to Ammospermophilus vocalisations by the criteria of TEMBROcCK (1968). These sounds were recorded and used to represent an out-group, termed “Sciuridae” (Fig. 8). Sciurus niger and Spermophilus (Otospermophilus) variegatus are anatomically relatively general- ised members of the famıly Sciuridae (BLAck 1963; Long and CaPpTAın 1974; WILSON 1960) and as such constitute a reasonable outgroup for analysıs of Ammospermophilus, for whom the sister group is unknown. Antipredator alarm trills occur in other ground squirrels in the holarctic subgenus Spermophilus (Spermophilus) which also inhabit new xeric habitats (BALPH and BALPH 1966; MELCHIOR 1971; BETTS 1976; LEGER et al. 1984). They also are present in the subgenus Spermophilus (Ictidomys) (Harrıs 1967; SCHwWAG- MEYER and BRown 1981). However, the trılls of these species are dissimilar to those of Ammospermophilus ın having longer pulse rates, higher pitch and no “kuk.” If the vocalisations of these taxa are used as an outgroup, the “kuk” element ıs left unanalysed. Five measurements from sonagrams of the calls of these squirrels (Tab. 6) are compared by use of cladıstic methods (HennıG 1979). These measurements are the salient varyıng characters in Ammospermophilus. The cladogram (Fig. 8) ıs rooted in the alarm calls of “Sciuridae” consisting of a “kuk” and a pure-toned, pulsed “buzz” (HorwicH 1972) or trill. It is difficult to determine whether quantitative characteristics are shared or different when mean values for all the taxa form a continuum. I estimated the confidence limits of all means by use of two standard errors averaged for the three large samples to determine overlap between samples. This method minimizes the effects of the disparate sample sizes. This approach is employed to determine relative primitive and derived character states. A multiple state transformation series (HENNIG 1979; BRETSKY 1979) is assumed to exist. Each significantly different (P =.05) mean ıs considered to represent a character state. Relative derived states, (Tab. 6, Fig. 8) are indicated by succeedingly larger number of superscripts and are based on the assumption that evolution is parsimoniously unidirec- tional. All Ammospermophilus difter from “Sciuridae” ın having a significantly more rapıd pulse rate, the trıll (MPD, e’-e’’’), and higher pitch (FMT, c’-c’’'). Character states e’ and c’ are synapomorphies (shared derived characters) of the genus. A. nelsoni shares no additional derived characters within Ammospermophilus and thus occupies the first branch within the genus. This species possesses an autapomorphy (unique derived character state) in MPD (e'’’). The remaining four species share a derived, higher-pitched call (FMT, c’’- c’'’). A. insularis, A. leucurus, and A. harrisii possess a derived higher frequency harmonic (FRH, d’). A. leucurus and A. harrisii are terminal sister taxa united by a highly derived long (DOC, a’), and pure-toned (FRO, b’’, or loss of “kuk”) vocalisation. These two species differ from one another only in their cascade ratio (CAR). There are only two homoplasies (convergence/reversal) in the tree: e’’ ın A. nelsoni and b’ in A. interpres. History of antipredator vocalisations of Ammospermophilus Comparisons with the vocalisations of other sciurids suggest that the short harsh calls of the geographically disjunct A. insularis, A. interpres, and A. nelsoni are primitive and that the long pure-toned trills ot A. leucurus and A. harrisü are derived. Because the changes inferred in vocalisation characteristics within the genus may have occurred facılely and rapidly, the shared derived similarities between A. lencurus and A. harrisii could be interpreted as having been 1. independently derived, or 2. inherited Ammospermophilus antipredator vocalısations 143 SeiunidaezaNElE INT INS BEI HAR Fig. 8. Cladogram based on multiple state transformation series of antipredator vocalisations of the five species of Ammospermophilus: A. nelsoni (NEL), A. interpres (INT), A. insularıs (INS), A. leucurus (LEU), and A. harrısi (HAR), using Spermophilus variegatus and Sciurus niger as the outgroup “Sciuridae”. Letters code character states as in Table 6 from a common ancestor which formerly occurred throughout their combined ranges. However, analysıs of skull morphology (BoLLes unpubl.) support the latter hypothesis; A. leucurus and A. harrisii are phenetically most similar to one another and are probably each other’s closest relatives. Ecological considerations suggest that A. insularıs, A. interpres, and A. nelsoni are adapted to closed habıtats and A. leucurus and A. harrisii to open desert conditions. These conclusions are consistent with what is known of the late Tertiary and Pleistocene history of the southwestern United States and northern Mexico. Thorn forest occurred through the late Tertiary (AxELRoD 1979) over much of what is today the Sonoran Desert, whereas the Mojave and most of the Great Basın were dominated by oak-pinon woodland (AxELROD 1979) until recent times. Table 6. Mean values of characters of alarm calls of Ammospermophilus and two other sciurid species Durationof “kuk” Fre- Pitch Frequency of Pulse Rate Call(DOC) quency Span (FMT) Harmonıc (MPD) Seconds (FRO) Kilohertz (FRH) Seconds Kılohertz Kılohertz Sciurus nıger 2: 0.133 | N=4 : Be Spermophilus varıegatus 1.600 | Ammospermophilus nelsoni 0.032 | e"" N22 Ammospermophilus interpres 1. : 0.063 | & N=3 Ammospermophilus insularıs 1. 0.044 N=4 Ammospermophilus leucurus 2. d' 0.052 | e" N=10 Ammospermophilus harrısı ! ; 0.050 N=18 144 Kathryn Bolles Late Wisconsin floras of the entire Southwest (from 22-17 thousand to 12-11 thousand ybp) (radıocarbon years before present) were pinon-juniper at mid-elevations (1525-5500 m) (MARTIN and MEHRINGER 1965; VAN DEVENDER et al. 1977; Raven and AxELRoD 1978; VAn DEVvENDER and Spauding 1979). The McKittrick tar pits which contain A. nelsoni (SchuLtz 1938) have a flora that indicates a relatively closed pinon-juniper woodland and chapparal (Pinus, Juniperus, Atriplex, Arctostaphylos) (Mason 1944). Packrat middens collected in the Mojave Desert dated at 10000 ybp contain predominantly pinon-juniper woodland species: Pinus, Juniperus, Fraxinus and Arctostaphylos (WELLS and BERGER 1967). The entrance of the present dominant species of all North American deserts, creosote (Larrea divarıcata), has been documented at less than 10000 ybp (Jornson 1976; WELLS and HUNnZIkER 1976). During full glaciations of the Pleistocene, desert floras were restricted to a narrow zone in the lower Colorado River valley and Gulf of California coast north of the tropic scrub (MARTIN and MEHRINGER 1965). Floristic changes 8000-4000 ybp suggest an Altitermal (Xerothermic) clımate (AxELROD 1966, 1979) during which desert species now separated by higher-elevation oak-pinon-juniper forest such as A. leucurus and A. nelsoni may have had continuous distributions (Munz 1974). There may have been continuous non-open habitat for populations of Ammospermophilus as late as 10000 ybp. Although the occurr- ence of the open-adapted A. harrisii and A. leucurus between the ranges of A. interpres, A. nelsoni and A. insularis could date from earlier times, the former two species may have originated with the expanding post-Wisconsin deserts. As recently as 4000 ybp, A.lencurus and A. nelsoni may have been in contact across the southern Sierra Nevada and Transverse ranges. The alarm vocalisation of A. lencurus and A. harrisii must be of very recent origin as are other behavioral, physiological, and anatomical adaptations for survival in an extremely xeric habitat (BARTHOLOMEW and Hupson 1959; Hupson 1962; KAvanAuU and RıscHER 1972; Maxson and MorToN 1974). Acknowledgements I thank N. E. Corıusas for his advice and for use of his laboratory facilities for the study of vocalisations; I am grateful also to R.L. Bezy, W.]J. Dixon, R. GOLDBERG, C. Jacos, M. GREENFIELD, and E. C. OLson for their interest and advice during this study, and especially grateful to ROBERT L. BEzy for his participation in the field. H. WALTER generously provided the Abstract in German. L. SmITH did the word processing. Permission to collect Ammospermophilus was kindly extended by Dr. LANDAZURI ORTIZ, Direccion General de la Fauna Silvestre de Mexico, and by the State of California Department of Fish and Game, Arizona Game and Fish Commission, and New Mexico Department ot Fish and Game. Financial assistance was provided by three research and field grants from the University of California Regents, and Office of Academic Computing. Zusammenfassung Zur Evolution und Variation der Warnrufe der nordamerikanischen Zieselgattung Ammospermophilus (Rodentia: Scinridae) Eine Analyse der Variation der Warnrufe nordamerikanischer Ziesel der Gattung Ammospermophilus zeigt, daf signifikante Unterschiede zwischen den Arten bestehen, während innerhalb einer Art nur geringe Unterschiede zwischen Individuen, Geschlecht, Jahreszeit und Umwelttemperaturen beste- hen. Das Ergebnis der Diskriminanzanalyse von zehn Lautparametern legt nahe, daß das durch- schnittliche Zeitintervall zwischen Lautelementen, die Schallfrequenz des energiereichsten Rufseg- ments und die des Rufanfangs am meisten zur Artentrennung beitragen. Der kurze und rauhe Triller von A. interpres, A. nelsoni und A. insularis scheint an Felsen- und Präriebiotope angepaßt zu sein, der lange und reine Triller von A. harrisi und A. leucurus dagegen an offene Wüstenbiotope. Vergleiche mit den Alarmrufen anderer Hörnchen lassen den Schluß zu, daß die kurzen Rufe primitiv sind und die Evolution der langen Triller eine Folgeerscheinung der rezenten Ausbreitung von Wüstenbiotopen in Nordamerika ist. Ammospermophilus antipredator vocalıisations 145 References AXELROD, D. I. (1966): The Pleistocene Soboba flora of southern California. Univ. Calıf. Publ. Geol. 60, 1-79. AXELROD, D. I. (1979): Age and origin of the Sonoran Desert vegetation. Calıf. Acad. Sci. Occ. Bapers, No. 132, 1-74. BALPH, D. M.; BarpH, D. F. (1966): Sound communication of Uinta ground squirrels. J. Mammal- ogy 47, 440-450. BARTHOLOMEW, G. A.; Hupson, J. W. (1959): Effects of sodıum chloride on weight and drinking in the antelope ground squirrel. J. Mammalogy 40, 354-360. BaPTIsTA, L. F.; PETRINOVICH, L. (1984): Socıal interaction, sensitive phases and the song template hypothesis in the white-crowned sparrow. Aniım. Behav. 32, 172-181. Bes, M. A.; HoFFMmAnN, R. $. (1977): Age determination and variation ın the red-tailed chipmunk, Entamias ruficandus. Murrelet 58, 26-36. BETTs, B. J. (1976): Behavior in a population of Columbian ground squirrels Spermophilus colum- bianus columbianus. Anım. Behav. 24, 652-680. BLack, C. C. (1963): Review of the North American Tertiary Sciuridae. Mus. Comp. Zool. Bull. 130, 109-248. BrAnD, L. R. (1976): The vocal repertoire of chıpmunks (genus Eutamias) ın California. Anım. Behav. 24, 319-335. BRETSsKYy, S. S. (1979): Recognition of ancestor-descendant relationships in invertebrate paleontology. In: Phylogenetic Analyses and Paleontology. Ed. by J. CRAcRAFT and N. ELDREDGE. New York: Columbia Univ. Press. pp. 113-163. BROUGHTON, W. B. (1963): Method in bioacoustic terminology. In: Acoustic Behavior of Animals. Ed. by R. G. Busner. New York: Elsevier Publ. Co. CALLAHAN, J. R. (1980): Taxonomic status of Eutamias bulleri. The Southwest Natur. 25, 1-18. CALLAHAN, J. R.; Davis, R. (1977): A new subspecies of the cliff chipmunk from coastal Sonora, Mexico. The Southwest Natur. 22, 67-75. CLARK, T. W. (1974): A field study of the ecology and ethology of the white-tailed prairie dog (Cynomys leucurus): with a model of the Cynomys evolution. Diss. Abstr. Intl. B 34. Cortias, N. E.; CoLLias, E. C. (1977): Weaverbird nest aggregation and evolution of the compound nest. Auk 94, 50-64. Cosens, S. E.; Farıs, J. B. (1984): A comparison of sound propagation and song frequency in temperate marsh and grassland habitats. Behav. Ecol. Sociobiol. 15, 161-170. Dixon, W.]J., (1981): BMDP Statistical Software, 1981. Berkeley: Univ. California Press. Dunrorop, C.; Davıs, R. (1975): Cliff chıpmunk vocalızations and their relevance to the taxonomy of coastal Sonoran chipmunks. J. Mammalogy 56, 207-212. FINDLEY, J. S.; Harrıs, A. H.; Wırson, D. E.; Jones, C. (1975): Mammals of New Mexico. Albuquerque: Univ. New Mexico Press. GANDER, F. F. (1967): Ground squirrels. “Picket pins” of our Wild West. Anım. Kingd. 70, 12-17. GREENE, H. W. (1978): Phylogeny, convergence, and snake behavior. Diss. Abstr. Intl. B 38 (9), no. 7802002. GREENE, H. W.; BURGHARDT, G. M. (1978): Behavior and phylogeny: construction in ancient and modern snakes. Science 200, 74-77. Hart, E. R. (1926): Changes during growth in the skull of the rodent Otospermophilus grammurus beecheyi. Univ. Calıf. Publ. Zool. 21 (14), 355404. Haraıs, J. P. W. (1967): Voice and associated behavior in Citellus tridecemlineatus and other ground squirrels. Diss. Abstr. Intl. B: 3101. HENNIG, W. (1979): Phylogenetic Systematics. 2nd printing of 1966 ed. Urbana: Univ. Illinois Press. HERBERT, H. (1983): Vocal communication in the megachiropteran bat Rousettus aegyptiacus: development of isolation calls during post-natal ontogenesis. Z. Säugetierkunde 48, 187-189. HOFFMEISTER, D. F.; Diersinc, V. E. (1978): Review of the tassel-eared squirrels of the subgenus Otosciurus. J. Mammalogy 59, 402—413. HorwIcH, R. H. (1972): The ontogeny of social behavior in the gray squirrel (Sciurus carolinensis). Advances in Ethology (Beihefte Z. Tierpsychol.) 8, 1-103. Hupson, J. W. (1962): The role of water in the biology of the antelope ground squirrel, Citellus leucnrus. Univ. Calıf. Publ. Zool. 64, 1-56. Josnson, H. B. (1976): Vegetation and plant communities of southern Calıfornia deserts — a functional view. In: Symp. Proc. Plant Communities of Southern California. Ed. by J. LArrınc. Calif. Native Plant Soc. Spec. Publ. 2, 125-164. Kavanat, J. L.; RıscHEr, C. E. (1972): Influences of ambient temperature on ground squirrel activity. Ecology 53, 158-163. KoeppL, ]. W.; HOFFMANN, R. $.; NADLER, C. F. (1978): Pattern analysıs of acoustical behavior in four species of ground squirrels. J. Mammalogy 59, 677-696. 146 Kathryn Bolles Kos, A. (1981): Entwicklung und Funktion der Ultraschallaute bei den Jungen von Myotis myotis und Wiedererkennung von Mutter und Jungem. Z. Säugetierkunde 46, 12-19. Kroopsma, D. E. (1982): Learning and the ontogeny of sound signals in birds. In: Acoustic Communication in Birds. Ed. by D. E. KroopsmA, E. H. MiLLEr, and H. OvUELLET. New York: Academic Press. Vol. 2, pp. 1-23. Kroopsma, D. E.; Bayrıs, J. R. (1982): Appendix: A world survey of evidence for vocal learning in birds. In: Acoustic Communication in Birds. Ed. by D. E. KroopsmA, E. H. MıLLErR, and H. OveLtert. New York: Academic Press. Vol. 2, pp. 311-337. LEGEr, D. W.; BERNEY-KEY, $. D.; SHERMAn, P. W. (1984): Vocalizations of Belding’s ground squirrels (Spermophilus beldingi). Anım. Behav. 32, 753-764. LEGEr, D. W.; Owıncs, D. H. (1978): Responses to alarm calls by California ground squirrels: effects of call structure and maternal status. Behav. Ecol. Sociobiol. 3, 177-186. Lewis, B. (1983): Directional cues for auditory localization. In: Bioacoustics, A Comparative Approach. Ed. by B. Lewis. London: Academic Press. pp. 233-260. Lewis, B.; Corzs, R. (1983): Sound localization in birds. Trends in Neuroscience 3, 102-105. Long, €. A.; CaPTAın, J. (1974): Investigation on the sciurid manus. I. Some new taxonomıc characters and their importance in the classification of squirrels. Z. Säugetierkunde 39, 98-102. MARLER, P. R. (1956): The voice of the chaffinch and its function as a language. Ibis 98, 231-261. MARSHALL, Jr., J. T.; MarsHauı, E. R. (1976): Gibbons and their territorial songs. Science 193, (4249), 235-237. MARTEN, K.; MARLER, P. (1977). Sound transmission and its significance for anımal vocalızation. I Temperate habitats. Behav. Ecol. Sociobiol. 2, 271-2. MarrIn, P. S.; MEHRINGER, Jr., P. ]J. (1965): Pleistocene pollen analysıs and biogeography of the Southwest. In: The Quaternary of the United States. Ed. by H. E. WRIGHT, Jr. and D. G. Frey. Princeton: Princeton Univ. Press. pp. 433-451. Mason, H. L. (1944): A Pleistocene flora from the McKittrick asphalt deposits of California. Calıf. Acad. Scı. (4 ser.) 25, 221-233. MAaTocHa, K. G. (1975): Vocal communication in ground squirrels, genus Spermophilus. Diss. Abstr. Intl. B 36 (6), 2663. Maxson, K. A.; Morton, M. L. (1974): Water and salt regulation in the antelope ground squirrel (Ammospermophilus leucurus). Comp. Biochem. Physiol. 47A, 117-128. MCcCARLEY, H.; McCARLEy, C. (1976): Canis latrans and C. rufus vocalızations: a continuum. The Southwest Natur. 21, 399-400. MELCHIOR, H.R. (1971): Characteristics of Arctic ground squirrel alarm calls. Oecologia (Berl.) 7, 184-1. Munz, P. A. (1974): A flora of southern California. Berkeley: Univ. Calif. Press. Nixousktı, A. A. (1974): [Geographic varıability of sound call rhythmic organızation in marmots of the group bobac (Rodentia: Scıuridae).] Zool. Zh. 53, 436444. Nikouskn, A. A. (1976): [Alarm call of marmots as a specific character.] Zool. Zh. 55, 1214-1224. NiKorskıı, A. A. (1979): [Species specificity of alarm calls in sousliks (Citellus, Sciuridae) of Eurasia.] Zool. Zh. 58, 1183-1194. NOTTEBOHM, F. (1972): Origins of vocal learning. Amer. Natur. 106, 116-140. Owıncs, D. H.; Hennessy, D. F. (1984): The ımportance of varıation in sciurid visual and vocal communication. In: The Biology of Ground-Dwelling Squirrels: Annual Cycles, Behavioral Ecology, and Sociality. Ed. by J. ©. MurıE and G. R. MICHENER. Lincoln: Univ. of Nebraska Press. pp. 169-200. PETERS, G. (1982): A note on the vocal behaviour of the Giant panda, Ailuropoda melanoleuca (David, 1869). Z. Säugetierkunde 47, 236-246. PETERS, G. (1984): On the structure of friendly close range vocalizations in terrestrial carnivores (Mammalıa: Carnivora: Fissipedia). Z. Säugetierkunde 49, 157-182. Raven, P. H.; AxELRoD, D. I. (1978): Origin and relationships of the California flora. Univ. Calıf. Publ. Botany 72, 1-134. RıEGER, 1.; PETERS, G. (1981): Einige Beobachtungen zum Paarungs- und Lautgebungsverhalten von Irbissen (Uncia uncia) im zoologischen Garten. Z. Säugetierkunde 46, 3548. Rogınson, J. W.; HOFFMANN, R. $. (1975): Geographical and interspecific cranial variation ın big- eared ground squirrels (Spermophilus): a multivariate study. Syst. Zool. 24, 79-88. ScHMIDLY, D. J. (1977): The mammals of Trans-Pecos Texas including Big Bend National Park and Guadalupe Mountains National Park. College Station: Texas A and M University Press. SCHMIDT, U.; JOERMANN, G.; SCHMIDT, C. (1982): Struktur und Variabilität der Wedhsssheniellkune juveniler Vampirfledermäuse Demon: rotundus). Z. Säugetierkunde 47, 143-149. SCHULTZ, J. R. (1938): A Late Quaternary Mammal Fauna from the tar seeps of McKittrick, California. Carnegie Inst. Wash. Publ. 487., Contr. to Paleontol. IV, 112-215. SCHWAGMEYER, P.; BRown, C.H. (1981): Conspecific reaction to playback of thirteen-lined ground squirrel vocalizations. Z. Tierpsychol. 56, 25-32. SHERMAN, P. W. (1977): Nepotism and the evoluionol alarm Calls Scienee 197, 1246-1253. Ammospermophilus antıpredator vocalısations 147 SIEBER, O©. J. (1985): Acoustic recognition between mother and cubs in raccoons (Procyon lotor). Behaviour 96, 130-163. SLOBODCHIKOFF, C. N.; COAST, R. (1980): Dialects in the alarm calls of prairie dogs. Behav. Ecol. Sociobiol. 7, 49-53. SMITH, C. C. (1978): Structure and function of vocalizations of tree squirrels (Tamiasciurus). J. Mammalogy 59, 793-808. STRUHSAKER, T. T. (1970): Phylogenetic implications of some vocalizations of Cercopithecus monk- eys. In: Old World Monkeys. Evolution, Systematics, and Behavior. Ed. by J. R. NArIEr and P. H. NaPIEr. New York: Academic Press. TEMBROcK, G. (1968): Land mammals. In: Animal Communication. Ed. by T. A. SEBEOR. Bloomington: Indiana Univ. Press. pp. 338404. VAN DEVENDER, T. R.; PHırLıps III, A. M.; MEaAD, ]J. I. (1977): Late Pleistocene reptiles and small mammals from the lower Grand Canyon of Arizona. The Southwest Natur. 22, 49-66. VAN DEVENDER, T. R.; SpauLDing, W. G. (1979): Development of vegetation and climate in the Southwestern United States. Science 204, 701-710. VAUGHAN, T. A. (1954): Mammals of the San Gabriel Mountains of California. Univ. of Kansas Publ. Mus. Natur. Hist. 7, 513-582. Weııs, P. V.; BERGER, R. (1967): Late Pleistocene history of coniferous woodland in the Mohave Desert. Science 155, 1640-1647. Weıts, P. V.; Hunzik&r, ]J. H. (1976): Origin of the creosote bush (Larrea) deserts of southwestern North America. Missouri Bot. Gardens Ann. 63, 843-861. Wiıscins, I. L. (1980): Flora of Baja California. Stanford Univ. Press. WıLey, R. H.; RıcHAarps, D. G. (1978): Physical constraints on acoustic communication in the atmosphere: implications for the evolution of anımal vocalızations. Behav. Ecol. Sociobiol. 3, 69-94. Wırson, R. W. (1960): Early Miocene rodents and insectivores from northeastern Colorado. Univ. Kansas Paleontol. Contr. Vertebrata 7, 1-92. ZELLEY, R. A. (1971): The sounds of the fox squirrel, Sciurus niger rufiventer. J. Mammalogy 52, 597-604. Author’s address: Dr. KATHRYN BoLLESs, Section of Birds and Mammals, Natural History Museum of Los Angeles, 900 Exposition Boulevard, Los Angeles, CA 90007 USA Z. Säugetierkunde 53 (1988) 148-161 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Robertsonian karyotype variation in the European house mouse, Mus musculus Survey of present knowledge and new observations By H. Wınking, BEATRICA Duuic, and G. BULFIELD Institut für Biologie der Medizinischen Universität zu Lübeck, Institute of Zoology, University of Zagreb, and Institute of Animal Physiology and Genetics Research, Roslin, Midlothian Receipt of Ms. 26. 1. 1987 Abstract The karyotype of the European long-tailed house mouse, Mus musculus, was studied from Zadar (Yugoslavia) and Ammoudıa (Greece). In mice from Zadar six and in those from Ammoudia three Robertsonian (Rb) chromosomes were present and have been identified by their G-band pattern. The composition of some of these marker chromosomes was unknown before, whereas some have been found to occur in other, geographically separated populations. The karyological details of both populations of the East-European distribution area were discussed together with observations on the karyotype of mice from Central and Western Europe and from Northern Africa. Fertility data of a feral mouse from Zadar, which was heterozygous for one Rb chromosome, show that this particular genotype does not necessarily imply impairment of fertility in wild mice, which is mainly present in Rb heterozygous anımals with a mixed genetic background of wild and laboratory mice. The Rb heterozygosity dependent impairment of fertility is due to segregational disorders during meiosis 1. Introduction Evans etal. (1967) and LEONARD and DECKNUDT (1967) were first to report the occurrence of a metacentric chromosome in laboratory mouse strains, and more findings of the same kind became known thereafter (see Gropp and WınkinG 1981). Similar changes, though with multiple pairs of metacentric chromosomes, were detected shortly later in wild mice from the Poschiavo valley in Southern Switzerland (Gropr et al. 1970), as well as from other areas of the Rhaetian Alps (GropPp et al. 1972). It became clear that these observations correspond to chromosome rearrangements ın a complex system of balanced Robertsonian (Rb) varıation, i.e. to changes due to centric translocations of two acrocen- tric chromosomes and commensurate reduction of the total number of chromosomes (MATTHEY 1966). The before mentioned and subsequent reports (CAPAnNA et al. 1976; v. LEHMANN and RADBruch 1977; Duuic et al. 1980; ApoLpH and Kreın 1981, 1983; GroPpP et al. 1982; Saıp et al. 1986) dealt with the occurrence, composition and characteristics of Rb metacentric chromosomes in mice obtained from local populations in the central (Lombardy, Switzerland, Germany), southern (Appennines, Sıcily), or western (Spain, Scotland) Europe and North Africa (Tunesia). Only little information exists, however, about the karyotype patterns of house mouse populations from eastern and southeastern Europe. It appeared worthwile to supplement the so far scarce cytogenetic data about this region by new and recent observations on mice from Yugoslavia and Greece, and to collect available, though scattered, karyogeographic data for an update comprehensive survey of the occurrence and distribution of Rb chromosome varıability ın the house mouse, including the derived laboratory strains. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5303-0148 $ 02.50/0 Robertsonian karyotype variation in the European house mouse 149 Material and methods Sixteen mice were trapped on two occasıons near Zadar (Dalmatıa), namely 10 animals in 1978 and sıx animals in 1981. A second site of investigation, about 640 kms southeast on the Adriatico-Ionian coast-line, is Ammoudıia (Epiros: Northeastern Greece) from where 19 mice were analysed in 1980. Karyotypes were established on Ilymphocyte metaphases from short time cultures (TRıMan et al. 1975) of peripheral blood cells. The blood samples were obtained by puncture of the retro-orbital sinus taken either from the original wild trapped mice (Zadar) or from their F-1 offspring wıth DBA/2 females (Greece). The identification of individual chromosomes was made on G-banded metaphases (SEABRIGHT 1971) using the criteria of the standard mouse karyotype presented by NEsBITT and FRANKE (1973). The designation of the Rb metacentric chromosomes and the arm composition followed the rules of the Committee on Standardized Nomenclature for Mice (1979). In addition, fertility of one natural male hybrid (Zadar) and of laboratory bred males derived from an interpopula- tion cross (Zadar x Poschiavo, see Table 4) was assessed by evaluation of matings with normal fertile laboratory female mice and by testis histology. Results Chromosome findings in mice from Zadar (Dalmatia) The cytogenetic findings, in particular the observation of Rb metacentric chromosomes and their arm composition, are recorded in Table 1. 15 among the 16 anımals showed a homozygous karyotype with sıx pairs of metacentric chromosomes and seven pairs of acrocentric autosomes plus the sex chromosome pair (Fig. 1). The remaining individual, which belonged to the 1978 sample, exhibited homozygosity for five pairs of Rb metacen- trics only, and heterozygosity for the Rb(8.17)38Lub metacentric, corresponding to a karyotype wıth an odd number of 11 metacentrics, 16 acrocentrics and the sex chromo- somes. Rb 33 Lub Tr Rb 38 Lub 1 15 12 14 13 17 1 5 6 10 9 8 E . a. = 4 ”. 2 3 4 £ 16 18 19 sw Fig. 1. G-banded karyotype ot a male mouse from Zadar (Yougoslavia) with six pairs of Rb chromosomes. Numbers refer to chromosomes of the standard karyotype of the mouse 150 H. Winking, Beatrica Dulic and G. Bulfield Chromosome findings in mice from Ammoudia (Greece) Table 1. Arm composition and designation of Rb translocations of mice trapped near Zadar (Yougoslavia) and Ammoudia (Greece) The karyotype of 19 wild trapped males was established by studying their 34 off- spring from crosses with DBA/2 females. In conventional chromosome preparations obtained from the F-1 progeny of 19 feral males, none, one or two Rb translocation metacentrics were found. However, from G-banding analyses ıt could be inferred that three different Rb translocations with a 3.10, 5.12 and 15.17 composition are pre- Rb translocations found in mice from Zadar! Ammoudia? Rb (1.11) 33 Lub Rb (5.15) 34 Lub Rb (6.12) 35 Lub Rb (10.14) 36 Lub Rb (9.13) 37 Lub Rb (15.17) 64 Lub Rb (3.10) 65 Lub Rb (5.12) 66 Lub Rb (8.17) 38 Lub sent in this population (Table 2). The cyto- genetic findings ın the analysed progeny of the 19 wild trapped males are summarized in Table 2. From the mode of transmission ' Populations highly homogeneous for Rb translocations: 15 among 16 animals were homozygous for all of the six Rb chromoso- mes (see text). - ” Population highly hetero- geneous for Rb translocations (see Table 2 and text) of the Rb metacentric chromosome into the F-1 descendants, it can be concluded that eight of the original males must have been heterozygous for one or more Rb translocations. Interestingly heterozygosity for all three Rb chromosomes of the investigated population must have occurred in at least one of the original males. This indicates that the Ammoudia mouse population is very heterogeneous. It seems possible that a few more Rb metacentric chromosomes are present in the small sample investigated, but were not transmitted to the limited number of karyotyped F-1 offspring. Table 2. Karyotype of first or second offspring of wild males (Ammoudia) with DBA/2 females Current no of wıld males first acrocentrics only acrocentrics only Rb (3.10) 65 Lub/+ Rb (3.10) 65 Lub/+ acrocentrics only acrocentrics only acrocentrics only acrocentrics only Rb (3.10) 65 Lub/+ VOOoONONWTRWDND jaN [@) acrocentrics only Rb (5.12) 66 Lub/+ Karyotypes of or second progeny Rb (3.10) 65 Lub/+ Rb (3.10) 65 Lub/+ Rb (3.10) 65 Lub/+ Rb (3.10) 65 Lub/+ acrocentrics only Rb (15.17) 64 Lub/+ Rb (15.17) 64 Lub/+ Rb (3.10) 65 Lub/+ and Rb (5.12) 66 Lub Rb (15.17) 64 Lub/+ and (Rb (15.17) 64 Lub Rb (3.10) 65 Lub/+ acrocentrics only Rb (3.10) 65 Lub/+ acrocentrics only acrocentrics only acrocentrics only Rb (3.10) 65 Lub/+ Rb (3.10) 65 Lub/+ Rb (3.10) 65 Lub/+ Rb (15.17) 64 Lub/+ Rb (3.10) 65 Lub/+ acrocentrics onl Rb (3.10) 65 Lub/+ acrocentrics only Karyogeographic evaluation of the findings in Zadar and Ammoudia Three of the Rb metacentric chromosomes found in Zadar (Dalmatıa), ı.e. (6.12), (10.14) and (8.17), show an arm composition identical t0 Rb chromosomes found in a mouse Robertsonian karyotype variation in the European house mouse 151 population near Palermo (v. LEHMANN and RApBrucH 1977), and two of them, (10.14) and (8.17), were shown to occur in mice from southern Germany (AnoLpH and KLEIn 1981). The Rb metacentric chromosome with the (5.15) arm composition occurs in a wide geographical range between Zadar (Dalmatia), central and northern Italy, Switzerland (CAPAnNA et al. 1976; GroPP et al. 1982) and Spain (ApoıpH and Krein 1981). Hence, the geographical distribution of this particular Rb metacentric chromosome follows an eastwestward direction. The (1.11) chromosome is likewise present in the northern African population (Monastir, Tunesia: Saıp et al. 1986), but the (9.13) chromosome of the Zadar population as well as two of the three Rb chromosomes of the Ammoudia (Greece) mice, i.e. (3.10) and (5.12) are unique among the Robertsonian chromosomes so far known from feral mouse populations (GRoPP and WInkınG 1981; BROOKER 1982; ADoLPH and KLEIN 1981; 1983), Fertility of the natural hybrid found in Zadar Heterozygosity of metacentrics of feral origin wıthin a laboratory mouse genome may cause meiotic anaphase I nondisjunction. This leads to the formation of unbalanced gametes which transmit the aneuploid genome into the zygote. As a consequence the litter size is reduced at a commensurate rate, since whole arm aneuploidies are subject to selective elimination during pregnancy (CATTANACH and MOoSsELEY 1973; GROPP et al. 1974). While such segregational disorders are mainly known from heterozygotes for Rb translocations introduced from feral mice into laboratory strains, no comparable know- ledge exists about Rb heterozygosity in a natural population. Therefore, the Zadar male with Rb(8.17)38Lub heterozygosity was mated several times with all acrocentric outbred NMRI females. The pregnant mothers were sacrificed at day 13 of pregnancy (plug day = day 1) and the numbers of corpora lutea and ımplants were recorded. Chromosome preparations were made from the fetal membranes of all alive implants. For comparison breeding records were established from males with the same Rb chromosome after introduction by five consecutive backcrosses into a laboratory mouse genome (NMRI) and from other Rb(8.17) heterozygous males, whose 8.17 Rb chromosome had a different source. One of these ıs the Rb(8.17)1lem chromosome detected by BARAnovV and DyBANn (1971) in a laboratory strain, and a second one ıs the Rb(8.17)6Sice chromosome isolated from a wild mouse of a Sıcilian population with a series of seven metacentrics (v. LEHMANN and RADBRUCH 1977). This latter chromosome has been isolated and is now carried on a mixed background of wild and laboratory derived genes. The results of the breeding experiments as shown in Table 3 provide no evidence for the existence of noteworthy Table 3. Effect of heterozygosity in males for Rb (8.17) translocations of different origin upon fertility' Rb chromosome no of percentage of pregnant corpora implants resorptions euploidd aneuploid lutea implants implants Rb (8.17) 38 Lub/+ 86 80 natural hybrid Rb (8.17) 38 Lub/+ laboratory mouse genome Rb (8.17) 6 Sic/+ 139 laboratory mouse genome Rb (8.17) 1 Iem/+ 144 laboratory mouse genome ! Data refer to observations on fetal progeny at day 13 sired by heterozygous fathers and all acrocentric mothers (NMRI). - * One embryo with 39 chromosome arms; presumably X0 152 H. Winking, Beatrica Dulic and G. Bulfield disturbances of meiotic segregation ın the Rb(8.17)38Lub/+ natural hybrid male, since the day 13 fetal progeny contains only a very low rate of resorptions. In addition there is no evidence for an increased rate of preimplantation losses. While the respective proportions of preimplantation losses are in the same low range in the progeny of both other Rb(8.17) heterozygotes, these show higher postimplantation losses than the Rb38Lub heterozygous feral male. Surprisingly the highest postimplantation losses, however, were found in crosses with Rb(8.17)38Lub/+ males, which carried the Rb chromosome in a laboratory mouse genome. The correspondence of the respective proportions of aneuploid MII plates in Rb1lem/+, Rb6Sic/+ and Rb38Lub/+ males with a laboratory mouse genome to the resorption rates lead us to assume that a major part of the resorptions is caused by chromosomal imbalance of the fetuses. In the Rb38Lub/+ feral male, even though MII evaluations were not possible, since the only available male died prior to the end of the breeding experiments, a correspondence of meiotic malsegregation rates and postimplanta- tion losses has to be postulated. Cytogenetic and reproductive analysis of laboratory hybrids bred from “Zadar” with multimetacentric type mice of different origin Breeding assays of anımals with newly identified Rb translocations against mice with a set of already known Rb chromosomes bear two advantages. Firstly, the type of pairing figures observed in 1st meiotic prophase can be used for examination on the correctness of the arm assignments of Rb translocations. Secondly, it provides information about the influence of complex Rb heterozygosity upon fertility. Both aims were borne in mind when males of the Zadar colony were crossed with females with the Rb1Bnr-7Bnr chromosome set (“tobacco mouse” - poschiavinus type). In case of proper identification of the “Zadar” metacentric chromosomes (see Tab. 1), the sıx “Zadar” and the seven “poschiavinus” metacentrics should be arranged in meiotic prophase of the hybrids in two chains of sıx and of five Rb metacentrics plus one ring bivalent of metacentric chromoso- mes. In fact, such pattern was observed in diakinesis figures. The male hybrids with such chains were sterile, and in the testis a severe manifestation of spermatogenetic arrest was observed ın histological sections suggesting a breakdown of spermiogenesis during meiosis or shortly thereafter. This finding supports observations reported by SEARLE et al. (1978), Evans (1976) and GroPP et al. (1982) that chain formation during prophase of meiosis I has a harmful effect upon the process of spermatogenesis. Discussion Survey of a present knowledge about Robertsonian variation in laboratory and wild caught mice By means of biochemical and morphological characteristics two groups within the species Mus musculus can be recognized in Europe (ZIMMERMANN 1949; THALER et al. 1981). Geographically one group consisting of the three subspecies “domesticus, brevirostris and praetextus” occupies areas west of a line running from the Baltic sea through Germany and down to the Black sea. The second group taxonomically designed as M. m. musculus is distributed east to the beforementioned borderline (see: ORSINI et al. 1983 and Fig. 2). Besides M. musculus, at least, two further Mus species exist in Europe occurring sympa- trically with one or the other of the M. musculus subspecies, namely M. spretus ın southwestern Europe and one Mus species in southeastern Europe, which ıs known as Mus 4 (THALER et al. 1981). A common karyological denominator to all these species are acrocentric chromosomes, and an all acrocentric karyotype can be considered as the Robertsonian karyotype variation in the European house mouse 153 Fig. 2. Geographical sites of house mouse populations with Rb chromosomes in Western Europe and Northern Africa. Characteristics of each population are given in Tab. 4. Solid line represents borderline of the western and eastern subspecies of M. musculus (see: Orsını et al. 1983) standard karyotype of the longtailed house mouse, M. musculus, as well as of the Asıatic subspecies of the house mouse. The presence of Robertsonian translocations ıs a special trait of the domesticus, brevi- rostris, praetextus group of M. musculus. Moreover, this karyotypic varıability seems to be limited to the European and African distribution area of house mice of this group. Altogether 22 different and characteristic Rb-constitutions have been found in the western part of Europe and Africa. The locations and the combinations of Rb-chromosomes present are listed in Fig. 2 and Table 4. Although “domesticus/brevirostris” type mice invaded large areas outside Europe along the routes of world traffic, at least few reports exist on the presence of Robertsonian translocations ın natural populations lıke ın Marıon Island (Rogınson 1978) far outside of Europe. In addition, two further reports about the occurrence of Robertsonian translocations in Asiatic house mice exist in the literature (CHAKRABARTI and CHAKRABARTI 1977; MORIWAKI et al. 1984). On the other hand, spontaneous Rb variation is known to occur in laboratory mice, whose biochemical and mitochondrial DNA characteristics have been shown to belong mainly to a “domesticus” background (YonEkAwAa et al. 1980; FERRIS et al. 1982). Altogether 22 different Rb metacentrics have been described in laboratory stocks (see: Gropp and WınkınG 1981; SEARLE and BEECHEY 1986). In contrast to wild mice all autosomes are involved in laboratory-type Rb translocations. In additoin, one case of an X-chromosome transloca- tion onto autosome 3 (ARROYO NOMBELA and RODRIGUEZ MurcıA 1977) has been described. In contrast, a Robertsonian Y-autosome translocation has never been observed. 154 H. Winking, Beatrica Dulic and G. Bulfield Table 4. Arm composition, designation and derivation of Rb-chromosomes in feral mouse populations of Europe and northern Africa Arm composition/ designation Rb (1.11) 33 Lub Rb (5.15) 34 Lub Rb (6.12) 35 Lub Rb (10.14) 36 Lub Rb (9.13) 37 Lub Rb (8.17) 38 Lub Rb (15.17) 64 Lub Rb (3.10) 65 Lub Rb (5.12) 66 Lub Rb (3.4) 1 Sie Rb (2.15) 2 Sic Rb (6.12) 3 Sic Rb (5.13) 4 Sic Rb (10.14) 5 Sic Rb (8.17) 6 Sic Rb (9.16) 7 Sic Rb (1.2) 18 Lub Rb (4.13) 19 Lub Rb (3.9) 20 Lub Rb (5.14) 21 Lub Rb (8.12) 22 Lub Rb (10.15) 23 Lub Rb (6.16) 24 Lub Rb (1.18) 10 Rma Rb (2.17) 11 Rma Rb (4.11) 12 Rma Rb (6.7) 13 Rma Rb (3.13) 14 Rma Rb (5.15) 15 Rma Rb (8.14) 16 Rma Rb (10.12) 17 Rma Rb (9.16) 18 Rma Rb (1.7) 1 Rma Rb (3.8) 2 Rma Rb (6.13) 3 Rma Rb (4.15) 4 Rma Rb (10.11)5 Rma Rb (2.18) 6 Rma Rb (5.17) 7 Rma Rb (12.14) 8 Rma Rb (9.16) 9 Rma Rb (1.2) 10 Lub Rb (5.13) 11 Lub Rb (3.9) 12 Lub Rb (4.17) 13 Lub Rb (6.16) 14 Lub Rb (8.14) 15 Lub Rb (10.12) 16 Lub Rb (11.15) 17 Lub Locality Zadar Am- moudia Palermo Island of Liparı Cittaduc- cale Ancarano Noof References locality (see Fig. 2) 1 this paper v. LEH- MANN and Rap- BRUCH 1977 GROPP and WINKING 1981 CAPANNA etal. 1976 CAPANNA etal. 1976 GROPP and WINKING 1981 Arm composition/ designation Rb (2.4) 25 Lub Rb (3.6) 26 Lub Rb (5.15) 27 Lub Rb (7.8) 28 Lub Rb (10.12) 29 Lub Rb (11.13) 30 Lub Rb (9.14) 31 Lub Rb (16.17) 32 Lub Rb (3.4) 39 Lub Rb (2.8) 40 Lub Rb (6.7) 41 Lub Rb (5.15) 42 Lub Rb (10.12) 43 Lub Rb (11.13) 44 Lub Rb (9.14) 45 Lub Rb (16.17) 46 Lub Rb (1.6) 47 Lub Rb (3.4) 48 Lub Rb (2.8) 49 Lub Rb (5.15) 50 Lub Rb (10.12) 51 Lub Rb (11.13) 52 Lub Rb (9.14) 53 Lub Rb (16.17) 54 Lub Rb (7.18) 55 Lub Rb (1.3) 1 Lub Rb (2.8) 2 Lub Rb (4.6) 3 Lub Rb (5.15) 4 Lub Rb (10.12) 5 Lub Rb (11.13) 6 Lub Rb (9.14) 7 Lub Rb (16.17) 8 Lub Rb (7.18) 9 Lub Rb (1.3) 56 Lub Rb (2.8) 57 Lub Rb (4.6) 58 Lub Rb (5.15) 59 Lub Rb (10.12) 60 Lub Rb (11.13) 61 Lub Rb (9.14) 62 Lub Rb (16.17) 63 Lub Rb (1.3) 1 Bnr Rb (4.6) 2 Bnr Rb (5.15) 3 Bnr Rb (11.13) 4 Bnr Rb (8.12) 5 Bnr Rb (9.14) 6 Bnr Rb (16.17) 7 Bnr Locality several places near Milano Gallarate Cremona Bergamo, Ardenno, Sondrio Sondalo Brusio, Sondalo No of locality (see Fig. 2) 8 References WINKING and GROPP 1983 GROPP and WINKING 1981 Robertsonian karyotype variation in the European house mouse Arm composition/ designation Rb (10.11) 8 Bnr Rb (4.12) 9 Bnr Rb (1.10) 10 Bnr Rb (2.14) 11 Bnr Rb (7.8) 12 Bnr Rb (13.16) 13 Bnr Rb (2.4) 67 Lub Rb (3.8) 68 Lub Rb (6.7) 69 Lub Rb (5.13) 70 Lub Rb (10.12) 71 Lub Rb (9.14) 72 Lub Rb (11.18) 73 Lub Rb (16.17) 74 Lub Rb (4.12) 1 Tu Rb (2.15) 2 Tu Rb (3.6) 3 Tu Rb (8.17) 4 Tu Rb (10.14) 5 Tu Rb (11.13) 6 Tu Rb (5.15) 16 Tu Rb (13.14) 17 Tu Rb (5.14) 18 Tu Rb (1.5) 19 Tu Rb (3.8) 20 Tu Rb (6.10) 21 Tu Rb (7.18) 22 Tu Rb (8.10) 23 Tu Rb (9.14) 24 Tu Rb (10.14) 25 Tu Rb (11.16) 26 Tu Rb (4.12) 1 Nam Rb (3.14) 7 Tu Rb (4.10) 8 Tu Rb (9.12) 10 Tu Rb (6.14) Locality several Alpine valleys, Bondo, Roveredo, Chiavenna Luino, Northern, Italy several places ın Southern Germany Belgium Orkney ıslands Table 4 (continued) Noof References localıty (see Fig. 2) 14 GROoPP etal. 1972 15 16 ADOLPH and Kein 1981; 1983 HÜBNER 1985 ÄADOLPH and KLEIN 1981; BROOKER 1982 17 18 Arm composition/ designation Rb (4.10) 8 Tu Rb (6.13) 9 Tu Rb (9.12) 10 Tu Rb (11.17) Rb (10.14) Rb (4.8) Rb (1.11) Rb (8.17) Rb (8.14) Rb (15.17) Rb (3.8) Rb (8.15) Rb (3.16) Rb (17.18) Rb (11.14) Rb (4.14) 11 Tu Rb (5.15) 12 Tu Rb (6.10) 13 Tu Rb (9.11) 14 Tu Rb (12.13) 15 Tu Rb (13.16) 1 Mpl Rb (3.12) 4 Mpl Rb (4.6) 5 Mpl Rb (5.14) 6 Mpl Rb (7.18) 7 Mpl Rb (8.9) 8 Mpl Rb (10.17) 9 Mpl Rb (13.15) 10 Mpl Locality 155 Noof References locality (see Fig. 2) Castle- town, Caithness places near Barcelona Ibiza Monastir 19 ADOLPH and KLEIN 1981; BROOKER 1982 ÄADOLPH and KLEIN 1981 BRITTON- DAVIDIAN 1983 Saıp etal. 1986 Some Rb chromosomes with similar arm composition as those observed in laboratory stocks were found in natural house mouse populations demonstrating that translocation events independently can occur involving the same chromosomes. Another point of interest is that some Rb’s of laboratory mice have been found in strains already containing a metacentric derived from a feral mouse (GroPpp and WınkinG 1981). This leads to the conclusion that the introduction of preexisting Rb metacentric chromosomes into the laboratory genome is not essential for the occurrence of Robertsonian changes in labora- tory strain. However, it may facılitate or enhance the generation of this type of chromoso- mal rearrangement. Considerably higher numbers of Rb translocation chromosomes have been found ın populations of wild living house mice. Together with the three Rb metacentrics from Dalmatıa and Greece described in this report the total number of Rb chromosomes with 156 H. Winking, Beatrica Dulic and G. Bulfield different arm composition amounts to 77. A lıst of update completeness recording all so far described Rb chromosomes using the current designation according to the arm composi- tion and the nomenclature rules (see: GroPP and WınkınG 1981) is shown in Table 5. 42 Rb metacentrics were observed in only one population or geographic site, whereas 35 Rb chromosomes were found at least twice and up to ten times with a mean occurrence of 2.1. Table 5. Composition and frequency of Rb translocations in karyotypically distinct populations of feral mice. Rb translocations of laboratory origin are marked by an asterisk Compo- Frequency Compo- Frequency Compo- Frequency Compo- Frequency sition sition sition sition DD mm wo MPrAarkeonNnoNnwmNND DD mo wWwwWwra Ana QeaD_DmArR Hana [N SI [8S) DA TS en En a BE nn u 5 oo NS) mWo—m Dre ea D rer be DnD DH |nNDNDMD 22 | 3 1 1 1 1 3 1 1 1 2 1 1 4 1 1 1 1 1 1 3 1 22 4 2 The most frequent and widely distributed Rb chromosomes are 5.15, 10.12, 9.14, 11.13 and 16.17 which were observed in seven or more different karyotypes. Sımilarly composed metacentrics in karyotypically distinct populations pose the question of whether they are derived by spread from one source or from multiple independent mutational events. Yet, the answer must remain open as long as definitive markers along the Rb chromosomes or genetic tests for the precise ascertainment of identity or diversity are not available. However, from the geographical distribution patterns ıt can be inferred that spreading of Rb chromosomes is in majority, though not all cases, a likely explanation (GroPpP et al. 1982). Even the presence of three sımilary composed metacentrics in the populations near Zadar (Yugoslavia) and Palermo (Sicily) might be attrıbutable to accidental introduction, e.g. by naval trade routes. It ıs possible that several new Rb chromosomes arose or accumulated independently after an initial spread of founder Rb metacentrics but a mutual spread between preexisting multimetacentric populations is similary possible. In this respect and with regard to spread by sea traffic, it is interesting to note that the population with metacentric chromosomes near Zadar is limited to a coastal area surrounded by populations with only acrocentric chromosomes (Duuid et al. 1980). The origin of mouse populations with Rb metacentric chromosomes outside Italy can be suspected to have Robertsonian karyotype varıation in the European house mouse 1157 occurred in coastal areas of Greece, Spain, Scotland and Tunesia suggesting accidental spread of Rb translocations into foreign populations vıa boat traffic. Interestingly Rb chromosomes are frequent in some parts of the British Isles (AporpH and Kreın 1981), which, as proposed by Davıs (1982), have been settled by continental mice via Vikings. It is likewise conceivable that the mutational events leading to the formation of biarmed chromosomes have taken place under the force of mixed genetic background of local and foreign genes. The finding of a “new” Rb chromosome ın a Japanese population that showed signs of contamination with European house mice, is in favour of this hypothesis (Morıwakt et al. 1984). Although Rb populations of the house mouse occur in many countries in Europe, the accumulation of Rb translocation chromosomes is most advanced in Italy. Altogether 11 populations with seven or more pairs of Rb chromosomes have been described. Similarly high numbers were only occasionally found outside Italy as in Tunesia. Under the viewpoint of time, the origin or introduction of Rb chromosomes in populations outside Italy is supposed to be more recent. This assumption is supported by the fact that the majority of populations wıth Rb translocations outside Italy, e.g. Greece, Spain, Southern Germany or Scotland is heterozygous or shows only lower numbers of Rb metacentrics. It has been claimed that Rb heterozygosity in natural populations plays an important role as a potent barrier and isolating mechanism between Rb and non Rb containing populations (WHITE 1978). This belief is based on the fact, that after introduc- tion of ferally derived Rb chromosomes into a laboratory mouse genome heterozygous Rb carriers show an impaired fertility (TETTENBORN and GRoPP 1970; CATTANACH and MoseL£y 1973; GroPpP et al. 1974). This characteristic has been attrıbuted to natural Rb- hybrids as well. The proof of unimpaired fertility of the natural hybrid of Zadar may indicate, that heterozygosity of at least some Rb chromosomes within their environmental genome does not have a negative effect upon fertility that is present after the introduction of the ferally devired Rb chromosome into a laboratory mouse genome (see Table 3). Since at the present time no real proof exists of a detrimental effect of Rb heterozygosity in natural hybrids, the role of chromosomes as an isolating mechanism and hence in the process of speciation has to be reconsidered. The great number of varıable composed Rb chromosomes in natural populations of the house mouse might lead to the assumption of a random involvement of all 19 acrocentric chromosomes of the mouse in the mutational process of Robertsonian centric transloca- tions. Yet, an evaluation of frequency of involvement of each one of the acrocentric elements of the mouse karyotype in the 77 known Rb metacentrics shows (Fig. 3) that some are very frequently represented, as e.g. chromosome 10, 12 and 14, whereas others like no 7 and 18 participate only occasıonally in the formation of Rb chromosomes, and chromosome 19 is, at least ın feral house mice, never involved as it is with the sex chromosomes. It follows that certain chromosomes are more susceptible for the hypotheti- cal mechanism which induces or facilitates the events of Rb translocations. If this assumption is correct, a sequential order of autosomes involved ın Rb translocations has to be postulated. Although there is no definitive way to assess the subsequent changes in Rb accumulating natural populations, the comparison of Rb population with low and high numbers of translocated chromosomes can provide some hints for the elucidation of this problem. For example, Rb chromosomes with autosomes nos. 7 or 18 involved are found mostly in populations with high numbers (8 to 9 pairs) of Rb chromosomes, whereas autosomes 12 and 10 as one arm of a Rb chromosome are represented almost in all Rb- populations, even in populations wıth low numbers of mutated chromosomes (1 to 3 pairs). Although the numbers of metacentric chromosomes of laboratory mouse strains are low, a similar nonrandom involvement of autosomes does not seem to exist. What may be the reason for the differential proneness of individual chromosomes of the wild mouse genome to be involved in centric translocations? MILLER et al. (1978) have pointed out that nucleolus organızer regions (NORs) do have some influence upon the 100 80 60 40 20 158 H. Winking, Beatrica Dulic and G. Bulfield 1 27 2737 74 5 6 ua a za 137 147157 16 TS SEGEN no of [\ oe \ N chromosomes Fig. 3. Percentage of each chromosome of the standard karyotype as one part of individual Rb translocations ın different populations of feral mice from Europe and Northern Africa. The calcula- tion is based on data of 22 mouse populations (see Tab. 4 and Fig. 2). Values above 100 % result from multiple involvement of certain autosomes in Rb heterogeneous populations (see Tab. 4, populations 15, 16 and 19). NOR bearing chromosomes are marked with an arrow frequency of Rb translocations, since in mouse cell lines chromosomes with active NORs are significantly overrepresented in Rb chromosomes. From laboratory mice it is known that chromosomes 12, 15, 16, 18 and 19 may bear NORs proximal to the centromeric heterochromatin. In metaphases of European wild mice silver NORs appear at the same chromosomes and locations, although the presence of silver NORSs is variable between feral mouse populations (Wınking et al. 1980). The high frequency of chromosomes 12 and 15 ın Rb translocations of wild mice supports the data of MILLER et al. (1978). However, autosomes 18 and 19 carry silver NORs as well, but are by far the least represented partners ın Rb translocations. This may indicate that other factors than NORs play the major role in the production of Rb translocations in feral mouse populations. Additional support of this view comes from the fact that Rb chromosomes with rRNA gene clusters on both sites of the centromere are lacking among the collection of Rb translocations in feral mice. Theoretically the only candidates could have been the combinations 15.17 and 16.17 of European wild mice, but the presence of NORs on chromosome 17 has only been documented in the Asıan subspecies M. m. molossinus (DEV et al. 1977). A further point, which might be of importance for the generation of Rb chromosomes, is the organization of the region intimately connected with the process of Rb rearrangement. From cytological observations it becomes clear that the breakpoints prior to translocation are located within the paracentromeric heterochromatin or C-band positive material (Gropp and WınkınG 1981). Heterogeneity of this region in respect of Robertsonian karyotype variation in the European house mouse 159 base pair composition has been ascertained between individual chromosomes of laboratory mice (THusT and Ronne 1980) and chromosomes of mouse cell lines (Marcus et al. 1980) as well as between species of the genus Mus (SEN and SHARMA 1980; BRown and DovErR 1980). Whether these differences are causally related to the different proneness of chromo- somes within the species Mus and between species of the genus Mus to undergo the process of Rb rearrangement is not yet clear and should be elucidated in further experiments. Acknowledgement In memoriam of Prof. Dr. ALFRED GRoPP, who discovered the Robertsonian polymorphism of feral mice and made possible these findings. Zusammenfassung Karyotypvariation durch Robertsonsche Translokationschromosomen bei der europäischen Hausmaus, Mus musculus. Eine Übersicht über den derzeitigen Wissensstand und neue Informationen Untersucht wurde der Karyotyp der langschwänzigen Hausmaus, Mus musculus, aus Zadar (Jugosla- wıen) und Ammoudia (Griechenland). In Zadar konnten sechs und in Ammoudia drei Robertsonsche (Rb) Chromosomen nachgewiesen und mit Hilfe der G-Banden identifiziert werden. Die Komposi- tion der Markerchromosomen war teils neu, teils in anderen, geographisch getrennten Populationen schon gefunden worden. Die karyologischen Befunde dieser beiden Populationen aus dem osteuropäi- schen Verbreitungsgebiet werden gemeinsam mit Beobachtungen zum Karoytyp mittel- und westeu- ropäischer sowie nordafrikanischer Hausmauspopulationen diskutiert. Die Fertilitätsdaten eines männlichen Wildfanges mit einer Rb-Heterozygotie lassen den Schluß zu, daß dieser Genotyp in Wildmäusen nicht mit einer Fertilitätseinbuße gekoppelt sein muß, wie sie in Wildmaus/Labormaus-Hybriden infolge von Meiosestörungen mit anschließender Bildung befruchtungsfähiger, aber aneuploider Gameten hinreichend dokumentiert ist. References ADOLPH, S.; KLEın, J. (1981): Robertsonian varıation in Mus musculus from Central Europe, Spain, and Scotland. J. Hered 72, 219-221. ADOoLPH, S.; KLEIN, J. (1983): Genetic varıation of wild mouse populations in Southern Germany. Genet. Res. 41, 117-134. ARROYO NOMBELA, ]. J.; RODRIGUEZ Murcıa, C. (1977): Spontaneous double Robertsonian translo- cation (Rb2.3) and Rb(X.3) in the mouse. Cytogenet. Cell Genet. 19, 227-230. BARANoOV, V. $S.; Dysan, A. P. (1971): Embryogenesis and peculiarıties of karyotype in mouse embryos with centric fusion of chromosomes (Robertsonian translocation). Ontogenez. 2, 164-176. BRITTON-DAaviDIan, J. (1983): Private communication. Mouse News Letter 69, 35. BROOkER, P. C. (1982): Robertsonian translocations in Mus musculus from N.E. Scotland and Orkney. Heredity 48, 305-309. Brown, $. D. M.; Dover, G. (1980): Conservation of segmental varıants of satellite DNA of Mus musculus in a related species: Mus spretus. Nature 258, 4749. CAPANNA, E.; GroPpp, A.; Wınkınc, H.; Noack, G.; Civiteru1, M.-V. (1976): Robertsonian metacentrics in the mouse. Chromosoma 58, 341-353. CATTANACH, B. M.; MoseL£y, M. (1973): Nondisjunction and reduced fertility caused by the tobacco mouse metacentric chromosomes. Cytogenetics 12, 264-287. CHAKRABARTI, $.; CHAKRABARTI, A. (1977): Spontaneous Robertsonian fusion leading to karyotype variation in the mouse. — First report from Asia. Experientia 33, 175. COMMITTEE ON STANDARDIZED GENETIC NOMENCLATURE FOR Mice (1979): New rules for nomen- clature of genes, chromosome anomalies and inbred strains. Mouse News Letter 61, 4-16. Davıs, S. J. M. (1982): Wild mouse morphometrics. Private communication. Mouse News Letter 66, 79-80. Dev, V. G.; TANTRAVAHI, R.; MILLER, D. A.; MiLLer, O. J. (1977): Nucleolus organızers in Mus musculus subspecies and in the RAG mouse cell line. Genetics 86, 389-398. Durie, B.; SoLparTovıic, B.; DUNDERSKI, Z. (1980): Distribution of karyotypes in Mus musculus Linnaeus, 1758 (Rodentia, Muridae) in some regions of Yugoslavia. Biosistematika 6, 203-210. Evans, E. P.; Lyon, M. F.; DacLısH, M. (1967): A mouse translocation giving a metacentric marker chromosome. Cytogenetics 6, 105-119. 160 H. Winking, Beatrica Dulic and G. Bulfield Evans, E. P. (1976): Male sterility and double heterozygosity for Robertsonian translocations in the mouse. In: Chromosomes Today. Ed. by P. L. PrArson and K. R. Lewis. New York: Wiley and Sons. Vol. 5, 75-81. FERRIS, $. D.; SAGE, R. D.; Wırson, A. C. (1982): Evidence from mtDNA sequences that common laboratory strains of inbred mice are descended from a single female. Nature 295, 163-165. GRroPP, A.; TETTENBORN, U.; v. LEHMANN, E. (1970): Chromosomenvariation vom Robertsonschen Typus bei der Tabakmaus, M. poschiavinus und ihren Hybriden mit der Laboratoriumsmaus. Cytogenetics 9, 9-23. GRoPP, A.; WINKING, H.; ZECH, L.; MÜLLER, H. J. (1972): Robertsonian chromosomal variation and identification of metacentric chromosomes in feral mice. Chromosoma 30, 265-288. GROPP, A.; GIERS, D.; KoLsus, U. (1974): Trisomy in the fetal backcross progeny of male and female metacentric heterozygotes of the mouse. Cytogenet. Cell Genet. 13, 511-535. Gropp, A.; Wınkınc, H. (1981): Robertsonian translocations: Cytology, meiosis, segregation patterns and biological consequences of heterozygosity. Symp. Zool. Soc. Lond. 47, 141-181. In: Biology of the House Mouse. Ed. by R. J. BERRY. London, New York: Academic Press. GroPP, A.; WINKING, H.; REDI, C.; CAPANNA, E.; BRITTON-DAVvIDIAN, J.; NoAck, G. (1982): Robertsonian karyotype varıation in wild house mice from Rhaeto — Lombardia. Cytogenet. Cell Genet. 34, 67-77. HüÜBNER, R. (1985): A Robertsonian translocation in wıld house mice from Belgium. Mouse News Betterr72,,1116-_ 11107: LEHMANN, E. v.; RADBRUCH, A. (1977): Robertsonian translocations in Mus musculus from Sicily. Experientia 33, 1025-1026. LEONARD, A.; DEKNUDT, G.H. (1967): A new marker for chromosome studies in the mouse. Nature, Lond. 214, 504-505. Marcus, M.; NIELSEN, K.; GoITEIN, R.; GroPpP, A. (1979): Pattern of condensation of mouse and chinese hamster chromosomes in G2 and mitosis of 33258-Hoechst treated cells. Exp. Cell Res. 122, 191-201. MATTHEY, R. (1966): Le polymorphisme chromosomique des Mus afrıcains du sous-genre Leggada. Revision generale portant sur l’analyse de 213 individus. Revue suisse Zool. 73, 585-607. MILLER, O. J.; MILLER, D. A.; TANTRAVAHI, R.; Dev, V. G. (1978): Nucleolus organizer activity and the origin of Robertsonian translocations. Cytogenet. Cell Genet. 20, 40-50. MORIWAKI, K.; YONEKAWA, H.; GOTOH, O.; MINEZAWA, M.; WıInkiIng, H.; GroPpPp, A. (1984): Implications of the genetic divergence between European wild mice with Robertsonian transloca- tions from the viewpoint of mitochondrial DNA. Genet. Res., Camb. 43, 277-287. NesgBITT, M. N.; FRANKE, U. (1973): A system of nomenclature for band patterns of mouse chromosomes. Chromosoma 41, 145-158. Oßsını, P. PH.; BONHOMME, F.; BRITTON-DAVIDIAN, J.; CROSET, H.; GERASIMOV, $.; THALER, L. (1983): The complex of species of genus Mus ın Central and Oriental Europe. II. Criteria for identification, distribution and ecological characteristics. Z. Säugetierkunde 48, 86-95. Rogınson, T. J. (1978): Preliminary report of a Robertsonian translocation in an isolated feral Mus musculus population. Mamm. Chrom. Newsl. 19, 84-85. Saıd, K.; JACQUART, T.; MONTGELARD, C.; SONJAYA, H.; HELAL, A. N.; BRITTON-DAVvIDIAN, ]. (1986): Robertsonian house mouse populations in Tunesia: A karyological and biochemical study. Genetica 68, 151-156. SEABRIGHT, M. (1971): A rapıd banding technique for human chromosomes. Lancett II, 971-972. SEARLE, A. G.; BEECHEY, C. V.; Evans, E. P. (1978): Meiotic effects in chromosomally derived male sterility of mice. Ann. Biol. Anım. Biochem. Biophys. 18, 391-398. SEARLE, A. G.; BEECHEY, C. V. (1986): Map of structural changes. Mouse News Letter 75, 17. SEN, $.; SHARMA, T. (1980): Quantitative varıation of “Mus musculus-like” constitutive heterochroma- tin and satellite DNA-sequences in the genus Mus. Choromosoma 81, 393-402. TETTENBORN, U.; GRoPP, A. (1970): Meiotic non-disjunction in mice and mouse hybrids. Cytogene- tics 9, 272-283. THALER, L.; BONHOMME, F.; BRITTON-DAVIDIAN, J. (1981): Process of speciation and semi-speciation in the house mouse. Symp. Zool. Soc. Lond. 47, 27—41. In: Biology of the House Mouse. Ed. by R. ]. Berry. London, New York: Academic Press. T#ust, R.; RonNnE, M. (1980): Localization of AT-clusters in normal mouse chromosomes by netropsin prefixation treatment ın vitro. Hereditas 93, 321-326. TRIMAN, K. L.; Davısson, M. T.; RODERICK, T. H. (1975): A method for preparing chromosomes from peripheral blood in the mouse. Cytogenet. Cell Genet. 15, 166-176. WHITE, M. ]J. D. (1978): Chain processes in chromosomal speciation. Syst. Zool. 27, 285-298. WINKING, H.; NIELSEN, K.; Gropr, A. (1980): Variable positions of NORs in Mus musculus. Cytogenet. Cell Genet. 26, 158-164. WInkIng, H.; Groprp, A. (1983): Multimetacentric wild mouse (M. musculus domesticus) popula- tıons. Mouse News Letter 68, 75. YONEKAWA, H.; MORIWARI, K.; GOTOH, O.; WATANABE, J.; HAYasHı, J.-1.; MiyAsHITA, N.; PETRAS, Robertsonian karyotype variation in the European house mouse 161 M. L.; TAGAsHIRA, Y. (1980): Relationship between laboratory mice and the subspecies Mus musculus domesticus based on restriction endonuclease. Jap. J. Genet. 55, 289-296. ZIMMERMANN, K. (1949): Zur Kenntnis der mitteleuropäischen Hausmäuse. Zool. Jb. (Syst.) 78, 301-322. Authors’ addresses: Dr. Heınz Wınking, Institut für Biologie der Medizinischen Universität zu Lübeck, Ratzeburger Allee 160, D-2400 Lübeck, FRG; Dr. BEATRıcA Duuid, Institute of Biology, University of Zagreb, Rooseveltov Trg 6, YU-41000 Zag- reb, Yugoslavia; Dr. GRAHAM BULFIELD, Institute of Anımal Physiology and Genetics Research, Edinburgh Research Station, Roslin, Midlothian, EH25 9PS, Scotland Z. Säugetierkunde 53 (1988) 162-169 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Sex differences and seasonal variation in habitat choice in a high density population of Waterbuck, Kobus ellipsiprymnus (Bovidae) By P. Wırtz and PETRA KAIsER Institut für Biologie I, Albert-Ludwigs-Universität, Freiburg, und Max-Planck-Institut für Verhaltensphysiologie, Seewiesen Receipt of Ms. 19. 1. 1987 Abstract Habitat use of waterbuck (Kobus ellipsiprymnus) was recorded during road transect counts in Lake Nakuru National Park, Kenya. Waterbuck were encountered significantly more often in open (riverine) forest and open grassland than expected from a random distribution through the available habitat types. There were significant differences in habitat use by different age and sex classes, with the lowest percent overlap (85 %) between adult females and young males. Seasonal differences in habitat use correlated with rainfall: with increasing rainfall waterbuck moved into open grassland, during periods of low rainfall waterbuck moved into open shrub. Introduction Waterbuck are antelopes the size of red deer; they occur throughout Africa south of the Sahara (Dorst and DAnDELOT 1970; HALTENORTH and DILLer 1977). The two subspecies Kobus ellipsiprymnus ellipsiprymnus and K. e. defassa interbreed in areas of geographical overlap, e.g. Nairobi National Park, Kenya (BackHAus 1958; KıLEy-WORTHINGTON 1965). Next to ımpala (cf. LEUTHOLD 1977; JARMAN 1979), waterbuck probably is the antelope species studied most intensively (references in SPINAGE 1982 and TOMLINSON 1980b, 1981; MELTON 1983). The population at Lake Nakuru National Park, Kenya, belongs to the subspecies K. e. defassa. With an average of 30 waterbuck per km?, regionally up to 100/km?, Lake Nakuru NP has by far the highest density recorded for waterbuck (KuTIL£k 1974; WIRTZ 1978). The high population density can be expected to have effects on the social behaviour of the species. In three earlier publications (WırTz 1981, 1982, 1983a) the social behaviour was described. In the following we present data on habitat utilization and on how it varies seasonally and between age and sex classes. Study area Lake Nakuru is a shallow alkaline lake at an altitude of 1760 m in the eastern Rift valley, about 130 km northwest of Nairobi, Kenya. Lake Nakuru has become famous as “the lake of a million flamingoes”’; up to 1.4 mill. flamingoes have been recorded there (VarEscHı 1978). To protect the spectacular avıfauna of the lake and its shorelines, the area was declared a bird sanctuary and later a National Park. A vegetation map of the park is given by Wırrz (1982, fıg. 1). Open grassland is dominated by Cynodon dactylon, Themeda triandra, Chloris gayana, and species of Andropogon and Hyparrhenia; near the lake shore the soda-resistant Sporobolus spicatus is abundant. Open and dense shrub is formed by patches of Tarchonantus camphoratus, Lantana trifolia, and Pluchea ovalıs. Open and dense forestes are formed by Acacıa xanthophloea, Olea africana and other Acacıa species (all species identifications based on CuURRY-LInDAHL [1971] and KagumaHo Karuvo [1980]). Both study years U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5303-0162 $ 02.50/0 Habitat choice of Waterbuck 163 were exceptionally wet years compared to long term means (KUTILER 1975); see fig. 2 for monthly rainfall values. The density and biomass of the larger mammals has been described by KurıLek (1974) and WIRTZ (1978, 1983b). A checklist including the rarer species was given by Wırrz (1979) and the group size frequencies of the eight most common antelope species was described by Wırrz and LÖRSCHER (1983). Waterbuck are the dominant larger mammal species in terms of numbers and in terms of biomass. Material and methods Field work lasted for 25 months, from November 1977 to December 1979. At the beginning of each month, all roads and tracks of the park were followed with a Landrover. Whenever an individual or a group of waterbuck was seen, the sex and age class of all members of the group and the habitat were noted on a dictaphone. A definition of the age and sex classes used (calves, subadult females, adult females, juvenile males, young males, adult males) is given in Wırrz (1982). Any waterbuck which could not be assigned to one of the age and sex classes (e.g. because it was partially hidden by the vegetation) was called “unidentified”. Counts were made from 7.00 am to 9.30 am and not during heavy rainfall; it took five to seven days to cover the whole park. The total length of the strip counted in this way was 114 km. Strip width differed according to vegetation. During three of the counts, the distance from the road to each group of waterbuck was estimated, which gives the average sight distance in the different types of habitat. To determine the relative areas of the different habitat types, a separate count was made: While driving the Landrover at a constant speed of 20 mph, the type of habıitat to the left and to the right was noted every 2 seconds. The proportion of each habitat along the counting strip multiplied with the average sight distance gives a measurement of the relative sizes of the areas: Table 1. Table 1. Relative size of the different types of habitat along the counted transect habitat type proportion along the counting average sighting relative size of counting strip (% of records) distance (m) area (% of total) open grass dense shrub open shrub dense forest open forest others For most of the varıables described in the “Results’ section, we have used the data for a “representative year” from August 1978 to July 1979 (n = 3017 groups, cumulative number of anımals recorded 17064). Results Population composition and an estimate of the variance produced by the counting method Table 2 shows the average population composition for the year August 1978 to July 1979. About a third of the population consisted of adult females. The sex ratio of adult and subadult anımals was not sıgnificantly different from 1:1 (507% 88:493% 2%; Table 2. Population composition (annual mean) adult young _juvenile adult subadult calves unidentified males males males females females n seen in the 12 counts 9259 3068 987 5353 1830 1055 1518 [0) /o (mean of the (OT 18:0 5.8 a oz 6.2 8.8 monthly percentages) 164 P. Wirtz and Petra Kaiser C[%AAF] 50 2 i I 20 79 2/79 3/79 4/79 579 6/79 W79 8/78 9/78 10/78 11/78 12/78 Fig. 1. Monthly varıation in the number of calves per hundred adult females p>0.2, t-test). The sex ratio of adult anımals was 1:1.6 (dd: 22). The percentage of calves in the population varied from 2.9 (June) to 9.7 (December); fig. 1 shows the numbers of calves per hundred adult females during the course of the year. How accurate are the monthly counts? A second count was begun immediately after the first one in June 1979 and in September 1979. Table 3 compares the results. Even though Table 3. Comparison of the results of two consecutive counts (percent of each age class ın the total) adult young juvenile adult subadult calves unidentified n males males males females females anımals 19.4 18.6 6.5 33.0 19.9 : : 1576 19:3 21.8 7.5 15 11.0 . 1282 16.0 23.4 53 34.6 8.2 i 5 1571 16.2 17.6 4.1 38.8 9.5 ! ; 1629 the absolute number of anımals counted during the first and second count can differ by as much as 300, the percentage of each age class in the total was very sımilar in the two counts made during the same month. The counting procedure apparently gives well-reproducible results. Habitat utilization General pattern Table 4 compares the relative frequencies of habitat types availabe (from table 1) and the proportions of waterbuck counted in them during the 12 months. Most of the habıtat covered during the counts consisted of open grassland and most waterbuck were seen in open grassland. However, waterbuck were not evenly distributed throughout the available habitat types: a larger proportion of waterbuck was seen in open grassland than expected for an even distribution (p<0.001, chi-square test). Habitat choice of Waterbuck 165 Table 4. Habitat availability and the proportion of waterbuck seen in the different types of habitat open dense open dense others shrub shrub forest forest proportion of anımals counted (n=17064) relative sıze of area \ DB 4.3 9.5 22 ZRH 4.5 15.3 0.7 Age and sex specıfic trends There were sex differences and age differences in the distribution of waterbuck: Table 5. The distribution of each age class differs with p<0.001 from an even distribution through the available habitat types (chi-square test). A considerably larger proportion of adult males was seen in open grassland than expected for an even distribution (compare Table 5. Percentage of different age classes and sexes in the different types of habitat (annual mean) young juvenile adult subadult calves unidentified females females open grass open shrub dense shrub open forest dense forest others n animals with lower part of table 4). In contrast, adult females were seen in open grassland less often than expected and in open forest much more often than expected. The difference in distribution pattern of adult males and adult females is significant wıth p<0.001 (chi- square test). A comparison of the different age classes shows how the sex-specific distribution pattern becomes more pronounced from calf to adult anımal. Calves (which remain hidden during the first two to four weeks of their life; SpinaGE 1982) were seen in open shrub and open forest more often than expected for an even distribution. The distribution of subadult females has shifted to an even stronger preoponderance in open shrub and open forest (comparison with calves: 0.10p p>0.05 for all anımals and p<0.05 for adult females alone, SPEARManN rank test); there was no discernible correlation between the amount of rain falling in a month and the percentage of animals counted in dense shrub, dense forest, Habitat choice of Waterbuck 167 and open forest. With decreasing rainfall the proportion of waterbuck counted in open grass decreased and the proportion of waterbuck counted in open shrub increased. Note that the second statement is not a necessary consequence of the first statement but a separate finding: animals leaving open grass areas could have dispersed into several other types of habitat. With increasing rainfall waterbuck moved from open shrub to open grassland, with decreasing rainfall waterbuck moved from open grassland to open shrub. There is no definite calving season in Nakuru waterbuck; newborn calves were seen throughout the year. The percentage of calves counted did, however, increase sharply during July to September (fig. 1). We tested whether adult females, in any month, changed habitat use ın a different direction than did adult males. While there were differences in the degree of change in habitat use, the directions of shift of adult females and of adult males resembled each other and corresponded to the pattern shown ın fig. 2. Discussion Waterbuck depend on permanent access to water; they almost completely lack the ability to reduce water loss in response to a shortage of water (TAayLoRr et al. 1969) and they tend to occur close to areas where water is readıly available (e.g. van LAVIEREN and Esser 1979; SPINAGE 1982; SINCLAIR 1985). In Nakuru National Park nearly all areas covered during the counts, including “open grassland”’, were within three kilometres’ distance to water and most of the “open forest”’ was forest along rivers draining into Lake Nakuru. Due to the exceptional amounts of rain during the study period, the waterholes were always full. Waterbuck groups are open groups; they continuously change their size and composi- tion. From a group in, for instance, open grassland some anımals may wander off into adjacent forest, while others from nearby open shrub may join those ın grassland. Individuals apparently make an independent choice and the number of anımals (instead of the number of groups) in a certain habıtat has therefore been used as a measure of habıtat utilization. Statements on habitat use of waterbuck in the literature are sometimes based on the proportion of anımals seen in different types of habitat but without relating these proportions to the relative areas available (e.g. SpınaGE 1982). In the most extensive study of antelope-habitat relationships, Hırst (1975) showed that the waterbuck of the Transvaal lowveld (South Africa) occured in riverine gallery forest more often than expected for an even distribution through available habitat types. No preference for open grassland was indicated in HırsT’s study. Caution must be used in the interpretation of such differences between study areas because the nature of the habitat described with the same name may differ in subtle ways which are nevertheless important to the anımals. Whereas most of the Nakuru grassland was close to water, the grassland in HırsT’s study area may not have been. Jarman (1972) and SıncLarr (1985) also found a preference of waterbuck for riverine forest and for grassland in the Zambezi valley and in Masai Mara National Park. Different visibility of the different age classes may have affected the counts and the apparent age differences could be an artefact. However, only for calves is this likely to be true: in dense habitat calves are more likely to be overlooked than larger anımals and this may have resulted in an overestimate of the proportion of calves in open habıtat. We assume that the pattern of habitat use shown by the anımals is adaptive, ı.e. that anımals using certain habitat types at a given time have a higher chance of survival and/or reproductive success. Food, predation and climate are the factors most likely to be of importance (but see PEEK et al. 1976 for an example of the influence of parasites). Differences in habitat utilization between age and sex classes can be due to differences in their requirements and/or to different action of external varıables. Seasonal differences can be due to changes in requirements and/or changes in external variables. 168 P. Wirtz and Petra Kaiser Except for occasıonal predation on calves by leopards, there was no predation on Nakuru waterbuck. Females spent most of their time inside territories (along the lake shore and along rivers) where the grass appeared to be greener and lusher than in open grassland further away from water (but no measurements of the nutritionally relevant components were taken). Females also spent more time feeding than did males (Wırrz and OLDEKOP, in prep.). Forage quality appeared to be the main factor governing the distribution of the females. The nutritional value of many forage species varıes greatly with the available moisture. Ungulates require on average 4-5 % crude protein in their diet, but the crude protein content of grass drops from about 8% ın the wet season to 1-3 % in the dry season (SIncLAIR 1975; AroLAYAN and FArunsHo 1978). Large scale migrations of ungulates, e.g. the spectacular migration of wildbeest, seem to be a response to forage quality and quantity. Small scale shifts in habitat utilization of nonmigratory African ungulates, such as those shown by Nakuru waterbuck, are probably an analogous process (KuTILEk 1979). “Choice” may be the wrong word to describe the process regulating the distribution of part of the male population. Territory holders excluded most of the remaining adult males and most of the young males from their territories (Wırrz 1982). Analysıs of faeces of waterbuck in Rhodesia (TomLınson 1980a, 1981) and the time budgets of the Nakuru waterbuck (WIRTZ and OLDEKOP, in prep.) indicate that bachelor males are relegated to nutritionally inferior areas. Probably, only a small portion of the adult male population was actually in the preferred location and the sex differences in habitat use were caused by the action of the territorial males that kept most of the male population out of the areas the females used for foraging. Differential distribution of males and females caused by the terrıtorial behaviour of some of the males has been described for several other antelope species and JARMAN and JARMAN (1973) suggested that, as a consequence, females benefit by an enhanced resource apportionment which might even lead to an increase ın calf production. Acknowledgements The data were collected while the senior author held a postdoctoral grant of the Max-Planck- Gesellschaft; the Kenyan Government gave research permit OP 13/7C245/4; as a voluntary field assistant, ANDREW BARTON assisted during several of the counts. Many thanks to them. Zusammenfassung Geschlechtsunterschiede und jahreszeitliche Variation in der Habitatwahl in einer Hochdichte- Population des Wasserbocks (Kobus ellipsiprymnus) Bei Transekt-Zählungen im Nakuru Nationalpark, Kenya, wurde registriert, wie häufig Wasserböcke (Kobus ellipsiprymnus) in den verschiedenen Habitaten gesehen wurden. Wasserböcke wurden signifikant häufiger in offenem flußnahem Wald und in offenem Grasland gesehen als dies bei einer gleichförmigen Verteilung über die vorhandenen Habitate zu erwarten war. Die Habitatnutzungsmu- ster verschiedener Altersklassen und der beiden Geschlechter waren signifikant voneinander verschie- den. Die geringste Überlappung in der Habitatnutzung (85 %) war zwischen adulten Weibchen und jungen Männchen. Die saisonalen Unterschiede korrelierten mit der Regenmenge: bei hohen Nieder- schlagswerten hielten sich die Wasserböcke häufiger ın offenem Grasland auf, bei niedrigen Nieder- schlagswerten mehr in offenem Buschland. References AFOLAYAN, T. A.; FAFUNSHO, M. (1978): Seasonal variation in the protein content and the grazing of some tropical savannah grasses. E. Afr. Wildl. J. 16, 97-104. BAcKHAus, D. (1958): Sind Kobus defassa (Ruppel, 1835) und K. ellipsiprymnus (Ogilby, 1833) verschiedene Arten? Säugetierkundl. Mitt. 6, 174. CURRY-LINDAHL, K. (1971): A short term ecological survey of Lake Nakuru and the surrounding area. Unpubl. UNESCO field science report. Dossr, J.; DANDELOT, P. (1970): A field guide to the larger mammals of Africa. London: Collins. HALTENORTH, T.; DILLER, H. (1977): Säugetiere Afrikas und Madagaskars. München: BLV Verlag. Habitat choice of Waterbuck 169 Hırst, S. M. (1975): Ungulate — habitat relationships in a South African woodland/savannah ecosystem. Wildlife Monogr. 44, 1-60. Jarman, M. V. (1979): Impala social behaviour: Territory-hierarchy, mating, and the use of space. Fortschr. Verhaltensforsch. (Z. Tierpsychol. Suppl.) 21, 1-92. Jarman, P. J. (1972): Seasonal distribution of large mammal populations in the unflooded middle Zambezi Valley. J. appl. Ecology 9, 283-299. JARMAN, P. J.; Jarman, M. J. (1973): Social behaviour, population structure and reproductive potential in impala. E. Afr. Wildl. J. 11, 329-339. KAGUMAHO Kakuyo, Y. B. (1980): The effect of translocation on the Rothchild’s Giraffes (Giraffa camelopardalıis rothchildi Lydekker) from Lewa Downs Farm to Lake Nakuru National Park. M. Sc. thesis, Univ. Nairobi. KıLEy-WORTHINGTON, M. (1965): The waterbuck (Kobus defassa Ruppell, 1835 and K. elhpsiprymnus Ogılby, 1833) in East Africa: spatial distribution and a study of sexual behaviour. Mammalia 29, 176-204. KUTILEx, M. J. (1974): The density and biomass of larger mammals in Lake Nakuru National Park (Kenya). E. Afr. Wildl. J. 12, 201-212. KuTiLek, M. ]J. (1975): Feeding strategies and foraging impact of non-migratory african antelope. Ph. D. thesis, Michigan State Univ. KUTILek, M. ]. (1979): Forage-habitat relations of non-migratory African ungulates in response to seasonal rainfall. J. Wildl. Managem. 43, 899-908. LAVIEREN, L. P. van; Esser, J. D. (1979): Numbers, distribution and habitat preference of larger mammals in Bouba Ndjda National Park, Cameroon. Afr. J. Ecol. 17, 141-153. LEUTHOLD, W. (1977): African Ungulates. Berlin: Springer Verlag. Merron, D. A. (1983): Populatin dynamics of waterbuck (Kobus ellıpsyprymnus) in the Umfolozi Game Reserve. Afr. J. Ecol. 21, 77-91. PEER, J. M.; ULricH, D. L.; MAckis, R. J. (1976): Moose habitat selection and relationships to forest management in North-eastern Minnesota. Wildlife Monogr. 48, 1-65. SINCLAIR, A. R. E. (1975): The resource limitation of trophic levels in tropical grassland ecosystems. J. Anım. Ecol. 44, 497-520. SINCLAIR, A. R. E. (1985): Does interspecific competition or predation shape the African ungulate community? J. Anım. Ecol. 54, 899-918. SPINAGE, C. A. (1982): A Territorial Antelope: The Uganda Waterbuck. Academic Press: London. TAYLoR, C. R.; SPINAGE, C. A.; Lyman, C. P. (1969): Water relations of the waterbuck, an East African antelope. Amer. J. Physiol. 217, 630-634. Tomıinson, D. N. S. (1980a): Seasonal food selection by waterbuck Kobus ellipsiprymnus in a Rhodesian game park. S. Afr. J. Wildl. Res. 10, 22-28. Tomuinson, D. N. S. (1980b): Aspects of the expressive behaviour of the waterbuck Kobus ellipsiprymnus ellipsiprymnus ın a Rhodesian game park. S. Afr. J. Zool. 15, 138-145. Tomuinson, D. N. S. (1981): Effects of the socıal organization of waterbuck Kobus ellipsiprymnus ellipsiprymnus (Ogılby, 1833) on forage-habitat utilization in a Rhodesian game park. Afr. J. Ecol. 19, 327-340. VARESCHI, E. (1978): The ecology of Lake Nakuru (Kenya). I. Abundance and feeding of lesser flamingoe. Oecologia 32, 11-35. WırTZz, P. (1978): Results of three game counts at Lake Nakuru National Park. Bull. EANHS 1978, 108-109. — (1979): A checklist of the larger mammals of Lake Nakuru National Park. Bull. EANHS 1979, 64-65. — (1981): Territorial defence and territory take-over by satellite males in the waterbuck, Kobus ellipsıprymnus (Bovidae). Behav. Ecol. Sociobiol. 8, 161-162. — (1982): Territory holders, satellite males, and bachelor males in a high density population of waterbuck (Kobus ellipsiprymnus) and their associations with conspecifics. Z. Tierpsychol. 58, 277-300. — (1983 a): Multiple copulations in the waterbuck (Kobus ellipsiprymnus). Z. Tierpsychol. 61, 78-82. — (1983b): The biomass of the larger mammals of Lake Nakuru National Park. Bull. EAHNS 5-7. WırTz, P.; LÖRSCHER, ]J. (1983): Group sızes of antelopes ın an East African National Park. Behaviour 84, 135-156. Authors’ address: Dr. PETER WIRTZ, PETRA KAIsER, Zoologisches Institut, Albert-Ludwigs-Univer- sıtät, Albertstr. 21a, D-7800 Freiburg Z. Säugetierkunde 53 (1988) 170-177 © 1983 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Summer food habits and quality of female, kid and subadult Apennine chamois, Rupicapra pyrenaica ornata Neumann, 1899 (Artiodactyla, Bovidae) By C. FERRARI, G. Rossı and C. CAvanı Istituto e Orto Botanico, Universita di Pavia, Pavia, and Istituto di Alimentazione Animale, Universita di Bologna, Bologna, Italy Receipt of Ms. 06. 01. 1987 Abstract Investigated the summer diet of a flock of the Apennine chamois comprising females, kids and subadults. Monthly observations were carried out between June and September from 1982 to 1984. The study area consisted of grasslands above the timberline belonging to the Festuco-Trifolietum thalii plant community, in the upper Val di Rose (Abruzzo National Park, Italy). Our data were obtained from direct observations of grazing anımals and from an analysis of the plants browsed. From June to September about 70 % of the total number of species are grazed. However the composition of the diet shows monthly shifts conditioned by the grassland phenology and the grazing selection. A tentative estimate of the main chemico-nutritional features of the diet suggests that the grazing selection keeps the Festuco-Trifolietum thalıi suitable to supply a protein-rich and fibre-poor diet during the whole summer. Introduction The quality and availability of food, together with security aspects, determine the habitat quality for the chamois (SCHRÖDER 1971; ELSNER-SCHACK 1985). While in the north-eastern chamois populations (Rupicapra rupicapra) this fact has been extensively studied (see Lovarı 1985 for a review), very little is known on the food ecology of the south western species (Rupicapra pyrenaica). In this paper we give some detailed data on the summer diet of females, kids and subadults of the Apennine subspecies of the south western chamoıs (Rupicapra pyrenaica ornata; NASCETTI et al. 1985). The only remaining population of this subspecies can be found in some mountains in the central Apennines of the Abruzzo National Park (Italy). Holocene remains and historical sources demonstrate a wider range in the central and southern Apennines up to historical times (Masını 1985). As for studies on the diet of the alpine chamois (e.g. ONDERSCHEKA 1974; DUNANT 1977; SCHRÖDER 1977) and of the Pyrenean one (BERDUCOU 1975; GARCIA-GONZALES 1984) these data will provide a background towards clarıfying the environmental require- ments of this “vulnerable” ungulate (THORNBACK 1980) and towards ensuring the success of possible reintroductions. Study area and methods The study area lies in the upper Val di Rose (Abruzzo National Park, Italy), between Mount Sterpalto (1966 m) and Mount Boccanera (1982 m). This area is part of the Camoscıara mountains which are the core of the chamois range in the National Park. Breaches formed by white and grey Dolomitic limestones dating from the Lias characterize the landscape (PRATURLON 1968). A mixed beach forest covers the slopes of the mountains up to about the timberline (1700-1800). According to the U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5303-0170 $ 02.50/0 Summer food habits and qualıty of Apennine chamois 1721 phytosociological approach (BRAuUn-BLANQUET 1964) the alpine grasslands belong to the vegetation types Festuco-Trifolietum thalii and Avena versicolor-Koeleria splendens community. Only the former is intensively grazed in summer by flocks of females, kids and subadults; the latter is grazed by solitary adult males. We therefore concentrated our observations on the Festuco-Trifolietum thalıi. The plant community was sampled with the phytosociological method (BRAuUn-BLANQUET 1964) and the cover of the species was estimated in a simplified way as 3: beyond 2/3, 2: from 1/3 to 2/3; 1 up to 1/3 of the minimal area of the releve. The sampling of the grazed species was carried out by direct observations. Our data were collected monthly for three years (1982 to 1984), from June to September, except for August (1983). The seasonal period we considered corresponds at first to the lactation and then (August-September) to the early weaning of kids. The flocks were observed using 12 x 50 binoculars from a distance of about 30-40 m. For each of the grazed species, the parts of the plant were noted (F: flowers; L: leaves; Fr: fruits), and the average grazing frequency (g.f.) was estimated (3: beyond 2/3; 2: from 1/3 to 2/3; 1: up to 1/3), using a slight modification of the scale proposed by Dunanr (1977). Nomenclature of taxa follows PıGnaTTi (1982), except for Graminaceae (Turin et al. 1964-80). A tentative estimate of the chemical composition of the diet was carried out by analysing 100 g samples of the monthly diet. The grazed parts of the plants were weighted on the basıs of species cover in the phytosociological relev&s. For each species and each month the calculation was as follows: Part grazing frequency X species cover Grazed part weight (g) = -100 Total grazing frequency x Total species cover The values we have considered in the calculation are in Table 1. Samples were weighed on a digital portable balance. They were sealed under vacuum in plastic containers and maintained at 0°C until analysıng. The samples were dried in a forced ventilation heater at a temperature of 65°C up to constant weight. The content of crude protein, calcıum, phosphorus, magnesium (A.O.A.C. 1984) and cell wall constituents (NDF; GoERInG and Van Soest 1970) were determined. Results The specific composition and phenology of diet The sampled species are listed ın Table 1. There are 54 grazed species; the total number of species ıs 78. Thıs means that, during the observation period, the flocks grazed 69.2 per cent of the total number of species. The specific composition of the grasslands is that of the vegetation type Festnco- Trifohietum thalıi, here characterızed by Festuca violacea macrathera, Trifolium thalı, Crepis aurea glabrescens and Plantago atrata. Considering as available only those species which, ın each month, were at phenological phases unlike “seedling” or “dried plant”, we can see (Fig. 1) that the number of available species is lowest in June, and is almost constant from July to September. Ratios between grazed species and available species show no significant differences during the whole summer. 18 70 60 Fig. 1. Number of grazed species (striped A > AR column) during the summer in relation to 3 40 = the total number of species (black col- a N umn) and the available species (white col- Ss = umn). Monthly ratios between grazed =) = = = and available species are the following: = > = = June: 0.57; July: 0.73; August: 0.66; Sep- 10 — = = — tember: 0.76. Grazed species are listed in A = = = = Table 1; available species in Table 2. e June July August September (Further explanations in the text) 172 C. Ferrari, G. Rossi and C. Cavani Table 1. Festuco-Trifolietum thalii Grazed species Species Aspidiaceae Polystichum lonchitis (1) Campanulaceae Phyteuma orbiculare (1) Chenopodiaceae Chenopodium bonus-henricus (1) Compositae Achillea millefolium (1) Adenostyles australıs (1) Aster bellidiastrum (2) Bellıs pusilla (1) Carduus carlinaefolins (1) Chrysanthemum tridactylites (1) Cirısum eriophorum (1) Crepis aurea glabrescens (3) Doronicum columnae (1) Leontodon hispidus (1) Senecio rupestris (1) Taraxacum officinale (2) Cruciferae Arabis alpına cancasıca (1) Cupressaceae Juniperus nana (1) Gentianaceae Gentiana Intea (1) Geraniaceae Geranium cinereum (1) Geranium macrorrhizum (1) Graminaceae Festuca dimorpha (1) Festuca nıgrescens (1) Festuca robustifolia (1) Festuca violacea macrathera (2) Phleum alpınum (1) Poa alpına (2) Juncaceae Luzula sieberi (2) Leguminosae Anthyllis vulneraria group (1) Astragalus depressus (1) Medicago lupulina (1) Trifolium pratense semipurpureum (1) Trıfolium repens (1) Trıfolium thalıi (3) Liliaceae Veratrum album lobelianun (1) Orobanchaceae Orobanche sp. (1) Plantaginaceae Plantago atrata (3) Plumbaginaceae Armeria majellensis (1) Polygonaceae Rumex acetosa (2) Primulaceae Soldanella alpina (1) DwreanNrD MN DW Summer food habits and qualıty of Apennine chamois 173 Table 1 (continued) Species Ranunculaceae Pulsatılla alpına alpına (1) Ranunculus apenninus (2) Rubiaceae Galium anisophyllum (1) Scrophulariaceae Linaria purpurea (1) Pedicularis comosa (1) Rhinanthus alectorolophus (1) Scrophularıa scopolii (1) Verbascum longıfolium (1) Thymelaeaceae Daphne mezereum (1) Umbelliferae Chaerophyllum hirsutum hirsutum (1) Chaerophyllum hirsutum magellense (1) Heracleum pyrenaicum orsinı (2) Pimpinella alpestris (1) Valerianaceae Valeriana montana (2) Violaceae Viola engeniae eugeniae (1) Mean cover (within brackets): 1=up to %; 2=from % to %; 3 = beyond %. Grazed parts: F=flowers; L = leaves; Fr = fruits. Grazing frequencies: 1=up to %; 2=from % to %; 3 = beyond %. Further explanations ın the text. Table 2. Festuco-Trifolietum thalii Available ungrazed species ın each month June (15) Compositae: Gnaphalium diminutum, Chrysanthemum tridactylites, Senecio rupestris, Taraxacum officinale; Cruciferae: Erysimum majellensis; Gentiana- ceae: Gentiana nivalıs; Labiatae: Stachys alopecurus divulsa; Thymus serpyllum group; Plumbaginaceae: Armeria majellensis; Primulaceae: Soldanella alpina; Ranunculaceae: Pulsatilla alpina; Rosaceae: Alchemilla nitida; Rubiaceae: Galinm anisophyllum; Scrophulariaceae: Scrophularıa scopolii; Umbelliferae: Pimpinella alpestrıs. July (12) Boraginaceae: Myosotis alpestris; Compositae: Achillea millefolum, Carduus chrysacanthus; Cupressaceae: Juniperus nana; Geraniaceae: Geranium cıne- reum, Geranium macrorrhizum; Labiatae: Lamium maculatum, Thymus ser- pyllum group; Primulaceae: Soldanella alpina; Rosaceae: Alchemilla nitida; Scrophulariaceae: Scrophularıa scopoli, Verbascum longifolium. August (15) Aspidiaceae: Polystichum lonchitis; Campanulaceae: Campanula scheuchzer:; Caryophyllaceae: Cerastium tomentosum; Convolvulaceae: Cuscuta sp.; Cupressaceae: Juniperus nana; Cyperaceae: Carex kitaibeliana; Graminaceae: Dactylis glomerata, Bromus erectus; Labiatae: Thymus serpyllum group; Rubia- ceae: Asperula arıstata, Galium anisophyllum; Scrophularıaceae: Linarıa purpu- rea, Verbascum longifolium; Umbelliferae: Seseli libanotis; Urticaceae: Urtica dioica. September (10) Aspidiaceae: Polystichum lonchitis; Caryophyllaceae: Cerastium tomentosum; Compositae: Achillea millefolium, Carduus chrysacanthus, Taraxacum offici- nale; Cupressaceae: Juniperus nana; Juncaceae: Juncus monanthos; Labiatae: Stachys tymphaea; Santalaceae: Thesium parnassi; Urticaceae: Urtica dioica. 174 C. Ferrari, G. Rossi and C. Cavani Nevertheless, if we compare two by two the monthly lists of grazed species (Table 1) through their common species and their species restricted to one month, we can see that the floristic composition of the diet shows a significant difference between June and July (x? = 3.94 with 1 df, and significant at P = 0.05). In Table 1 we also can see that: only the flowers of the species Phyteuma orbiculare, Achillea millefolium, Cirsium eriophorum, Senecio rupestris, Anthyllis vulneraria group, Galinm anisophyllum, Linarıa purpurea, Valeriana montana, Viola eugeniae subsp. eugeniae are eaten. 9 species altogether. 50 40 31 8 30 =2 = ö — == — 2 20 = = = Fig. 2. Number of grazed species ö = = = (T) subdivided according to parts = — — eaten. F= flowers; Fr= fruits; 0 — = = L = leaves Tre TFL TF-rL July August September Only the leaves are grazed in the following species: Polystichum lonchitis, Gentiana Iutea, Geranium macrorrhizum, Festuca dimorpha, Festuca violacea subsp. macrathera, Phleum alpinum, Medicago lupulina, Juniperus nana, Pedicularis comosa, Verbascum longifolium, Daphne mezereum. 11 species ın all. The other grazed species (34) are subjected to an indiscriminate form of grazing. Fig. 2 shows that in 3 July there ıs a clear preter- 4 60 a n = = ence for flowers. From = T z . d.m. R Y R July to August there ıs a 50 x 5 =} : 2 z = marked decrease in the 40 grazing of flowers. (x? = © 5 s 30 & R 0 2 18.50 with 1df, highly a) A je ® significant). Leaves are 20 ö a a = IN E grazed with high frequen- 10 Nah ni a a cy during the whole obser- a N hl ls ae vation period 1.25 ö © S PS ® . . ka z S 8 8 The main chemico-nutri- 5 hi [e} . . 0.75 2 Se ® E tional features of the diet Zr © [ex a N . 0,50 nn Erz 3 Fig. 3 shows the results of Ole E & our tentative estimate of | _ Ip 025 u some chemico-nutritional June July August September features of the diet. The Fig. 3. The main chemico-nutritional features of the summer diet. differences between our Monthly per cent content on the dried matter (d.m.) of crude monthly data have no protein (CP), cell wall constituents (NDF: neutral detergent fiber) statistical significance. and some elements. (Further explanations in the text) However, their apparent trends suggest a decrease in the crude protein (CP) and phosphorus content of the diet and an increase in cell wall constituents (NDF), calcıum and magnesium, from June to September. 0.00 Summer food habits and quality of Apennine chamois 11778 Discussion The grassland grazed in summer by females, kids and subadults of the Apennine chamois belong to a vegetation type (Festuco-Trifolietum thalıı) which is common in the alpine vegetation belt of the Alps, but ıs rare and extrazonal ın the high mountains of northern and central Apennines (PıcGnarTı 1979). Some significant subspecies such as Festuca violacea subsp. macrathera and Crepis aurea subsp. glabrescens indicate a slight chorologi- cal difference between the Apennine community and the Festuco-Trifolietum described for the Alps (BRAUn-BLANQUET 1949-50). As in the Alps its distribution is restricted to sites with a long-lasting snow covering and weakly acid soils. In the Abruzzo National Park, such environmental conditions occur in some northern slopes of the Camosciara moun- tains — which include the upper Val di Rose - but are rare and fragmentary elsewhere, as, for instance, on Mount Amaro (1862 m). The striking difference between the great number of chamois in the Camosciara with respect to other Park areas could be explained on the basis of the importance of the Festuco- Trifohietum thalıi as a food source especially during lactation. The phenology of this plant community has a role in determining the grazing habits of females and subadults and some nutritional shifts in the diet. From June to September the increase of magnesium (from leaves) and the decrease of phosphorus (from flowers), as our data indicate, are well known phenologically related facts. The high values found in June in the phosphorus con- tent of the diet (0.42 % of the o d.m.) may be related to the 0.25 corresponding chamois pre- ° 020 ference for the floral parts of 3 plants. Moreover, ıt is well - known that grasses and many other plants show decreasing * °% phosphorus contents evolving 2 eo from the early stage (0.40% SZ of the d.m.) to the end of ° 010 r flowering (0.20% of the 2 d.m.; BouQauET and Gur- z GUEN 1979). 2 y = 0.06x -0.04 Onwsrhesscontrary, ‚the, g °°° phenologic decrease of pro- ® ® teins (from young leaves and flowers) corresponds to only 0.00 an apparent slight decrease of June July August September protein content in the diet. As by Kie, 4, this fact = 4. a = ar 2 Bene frequency of guminosae leaves and the grazing frequency of leaves of other may be well explained on the families. Grazing frequencies (g.f.) are estimated according to basis of an increasing grazing Dunant (1977), modified. June: 0.03; July: 0.10; August: 0.13; frequency of Leguminosae September: 0.23. The correlation coefficient r = 0.98, with 1 df, leaves from June to Sep- is highly significant tember. The proteins of these leaves have a high digestibility and exercise a stimulating effect on the rumination (GRENET and DEMARQuILLY 1977). In Trifolium repens ULyATT (1981) observed a good digestibility even in an advanced vegetative state, as the herbivores only graze the leaves and the leaf- stalks of this clover. This removal should provoke a regeneration of a new leaf tissue which would keep up the supply of low fibre content forage but with high protein content. 176 C. Ferrari, G. Rossi and C. Cavani A sımilar situation could come about in the Festuco-Trifolietum thalıı, where Trifohum thalıi and other Leguminosae are abundant and are intensely browsed (see Table 1). In the same seasonal period the apparent increase ın calcıum content can be accounted for in the same way, ı.e. on the basıs of the increased browsing of the leaves of the Leguminosae (Van SozsT 1982). Finally, the diet presents fairly constant fibre values, at least up to the whole of August. The anımals’ careful grazing selection may have contributed to this phenomenon, as they choose the less fiıbrous parts of plants. This last aspect has been highlighted in the alpine chamoıs (DRESCHER-KADEN 1981). Our results based on the analysıs of a diet reconstructed by browsing data, need a confirm from more accurate investigations. However they suggest for the Apennine chamois what is already known in general for other ruminants (ArnoLD 1981): the grazing selection tends towards obtaining protein-rich diets which are however poor in fibres. In conclusion, two results of our studies are to be emphasized: 1. In the late spring (June), summer (July-August) and early autumn (September), the females, kids and subadults of the Apennine chamois seem to depend mainly on a rare and extrazonal plant community in the Apennines, such as the Festuco-Trifolietum thalıı. 2. The chamois grazing selection keeps this vegetation type suitable to supply a protein- rich diet in a seasonal period which corresponds to the lactation and to the early weaning of the kids. Acknowledgements We wish to thank the Direction of the Abruzzo National Park for providing practical facilities in the course of our research. We are also grateful to $. LovaRl for the critical reading of the manuscript. The research was partly supported by a grant from the Italian Ministry of Education. Zusammenfassung Beschaffenheit und Qualität der Sommernahrung von Weibchen, Kitzen und Subadulten der Apennin- gemse, Rupicapra pyrenaica ornata Neumann, 1899 (Artiodactyla, Bovidae) In den Monaten Juni bis September 1982-1984 wurden im oberen Val di Rose (Abruzzen-National- park, Italien) an einer Herde von Apenningemsen (Weibchen, Kitze, Subadulte) monatliche Beobach- tungen durchgeführt, um über dıe Sommernahrung Aufschluß zu erhalten. Die Vegetation im Untersuchungsgebiet oberhalb der Baumgrenze besteht aus Weiderasen (Festuco-Trifolietum thalıı). Unsere Ergebnisse basieren auf Direktbeobachtungen der äsenden Gemsen sowie auf Analysen der aufgenommenen Pflanzen. Von Juni bis September werden ungefähr 70 % der vorhandenen Pflanzen- arten abgegrast, doch die Phänologie des Weiderasens hat großen Einfluß auf die Nahrungswahl. Abschätzungen von chemischer Eigenschaft und Nährwert der Nahrung weisen darauf hin, daß das Festuco-Trifolietum thalii sich durch selektives Abweiden als geeignet erweist, eine proteinreiche und faserarme Nahrungsquelle während des ganzen Sommers zu liefern. Literature A.O.A.C. (1984): Official Methods of Analysis of Association of Official Analytical Chemists. 14th ed. Arhigton, U.S.A. Arnotrn, G. A. (1981): Grazing behaviour. In: World Animal Science. B 1: Grazing Animals. Ed. by F. MorL£Ey. Amsterdam: Elsevier. pp. 79-104. BERDUcOU, C. (1975): L’alimentation hivernal de l’isard. Bull. Off. Nat. Chasse, Special, 4, 195-227. BoucaueT, $. P.; GUEGUEN, L. (1979): Constituents mineraux majeurs. In: Prevision de la valeur nutritive des aliments des ruminants. Ed. by C. DEmArQuıLLy. Versailles: INRA. pp. 189-202. BRAUN-BLANQUET, J. (1948-50): Übersicht der Pflanzengesellschaften Rätiens. Vegetatio 1, 29-40; 129-146; 285-316; 2, 10-37; 215-237; 341-360. BRAUN-BLANQUET, J. (1964): Pflanzensoziologie. Grundzüge der Vegetationskunde. 3. Aufl. Wien: Springer. DRESCHER-KADEN, U. (1981): Vergleichende Untersuchungen zur Nahrungswahl von Gams- und Rotwild unter besonderer Berücksichtigung der mikrobiellen Besiedlung und der Verdauungsvor- gänge im Pansen. Fortschr. Tierphysiol. und Tiernahrung 12, 1-108. DunAaNnT, F. (1977): Le Regime Alımentaire du Chamois des Alpes (Rupicapra rupicapra rupicapra Summer food habits and qualıty of Apennine chamoıs 177 L.): Contribution Personelle et Synthese des Donnees Actuelles sur les Plantes Broutees. Revue Suisse Zool. 84, 883-903. ELSNER-SCHACK, I. von (1985): Seasonal changes in the sizes of chamois groups in the Ammergauer Mountains, Bavaria. In: The Biology and Management of Mountain Ungulates. Ed. by S. LovaRrı. London: Croom-Helm. pp. 148-153. GARCIA-GONZALES, R. (1984): Comparaciön de la dieta estival entre sarrıos jovenes y adultos. Acta biol. mont. 4, 333-340. GOERING, H. K.; SogsTt, P. J. van (1970): Forage Fiber Analyses. Agr. Handb. No. 379. Washington, UL GRENET, E.; DEMARQuILLY, C. (1977): Utilisation de l’azote des fourrages verts par le mouton en croissance; influence du stade de vegetation, de l’espece fourragere, de la fertilization azotee et de l’addition d’orge. Ann. Zootech. 26, 481-501. Lovarı, S. (ed.) (1985): The Biology and Management of Mountain Ungulates. London: Croom- Helm. Masını, F. (1985): Würmian and Holocene chamois of Italy. In: The Biology and Management of Mountain Ungulates. Ed. by $. Lovarı. London: Croom-Helm. pp. 3144. NASCETTI, G.; LOVARI, S.; LANFRANCHI, P.; BERDUCOU, C.; MATTIUCCI, $.; Rossı, L.; BuLLinı, L. (1985). Revision of Rupicapra Genus. 3. Electrophoretic Studies demonstrating Species Distinc- tion of Chamois Populations of the Alps from those of the Apennines and Pyrenees. In: The Biology and Management of Mountain Ungulates. Ed. by S. Lovarı. London: Croom-Helm. pp. 56-62. ONDERSCHERA, K. (1974): Ernährungsprobleme beim Gamswild. Tag. Ber. 1 int. Gamswildtreffen, Oberammergau, pp. 34-52. PıGnarTTı, $. (1982): Flora d’Italia. 3 vols. Bologna: Edagricole. PRATURLON, A. (1968): Note illustrative della Carta Geologica d’Italia. Foglio 152 (Sora). Servizio Geologico d’Italia, Napolı. & SCHRÖDER, W. (1971): Untersuchungen zur Okologie des Gamswildes (Rupicapra rupicapra L.) ın einem Vorkommen der Alpen. Z. Jagdwiss. 17, 113-168; 197-235. SCHRÖDER, W. (1977): Räumliche Verteilung und Nahrungsauswahl von Gams und Rotwild im Hochgebirge. Forstw. Cbl. 96, 94-99. SoEST, P. J. van (1982): Nutritional Ecology of the Ruminants. Corvallis, USA: ©. & B. Books. THORNBACK, J. (compiler) (1980): A Draft Community List of Threatened Species of Wild Fauna and Flora. Manuals. 2 vols. E.C.P.S., Commission of the European Communities. Mimeographed. Turin, I. G.; Heywoop, V. H.; BurGzs, N. A.; MOORE, D. M.; VALENTINE, D. H.; WALTERSs, $S. M.; Wesc, D. A. (1964-80): Flora Europaea. 5 vols. Cambridge: Cambridge University Press. ULyATT, M. J. (1981): The Feeding Value of Temperate Pasture. In: World Animal Science. B 1. Grazing Animals. Ed. by F. Morr£y. Amsterdam: Elsevier. pp. 125-139. Authors’ addresses: C. FERRARI and G. Rossı, Istituto e Orto Botanico, Via $. Epifanio 14, I-27100 Pavia, Italy, and C. Cavanı, Istituto dı Alimentazione Animale, Via S. Giacomo 11, 1-40126 Bologna, Italy Z. Säugetierkunde 53 (1988) 178-187 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 An annual rhythm in reproductive activities and sexual maturation in male Japanese serows (Capricornis crispus)"? By T. Tısa, M. SaTo, T. Hırano, I. Kıra, M. SuGımura and Y. Suzuki Department of Theriogenology, Faculty of Agriculture, Gifu University, Japan Receipt of Ms. 2. 1. 1987 Abstract Changes in reproductive activities were examined in male Japanese serows captured in December to March, 1982 to 1985, in Gifu Prefecture. Examinations were conducted on the spermatogenic activity, testosterone and androstenedione levels in tissue as well as in serum and fructose concentration in the seminal vesicle. In fully adult males over 2.5 or 3 years of age, the highest values in every examination were nearly always found in December. From December to March, the values showed a decided tendency to decrease. A highly significant correlation was found, respectively, between the sper- matogenic activity and the testosterone level, and between the fructose content and the testosterone level. These results indicate the presence of an annual rhythm in the reproductive activity in males of this species. In young anımals, the spermatogenic activity, testosterone levels, and fructose contents progressed increasingly with age and attained the adult levels at the age of 2.5 to 3. The concentration of androstenedione in serum of youngs showed the same level as in adults, and there was no definite tendency to decrease associated with sexual maturation. Introduction Until our first published study on spermatogenesis in Japanese serows (TıBA et al. 1981a, b), there had been no information available concerning reproductive physiology in males of this species, though it had been known from a few works on sexual behaviour in males, that mating occurs most frequently in October and November (AkasakA 1979; AKASAKA and MaruyaMmA 1977). As for females, several important facts had been collected about reproductive activities (ITO 1971; Komorı 1975): A single young was born usually from March to June, sometimes in September but seldom in October, the first parturiton in adult females occurred usually at 3 years of age; and the gestation period was 211-213 days in captıvıty. From our previous studies mentioned above, it was strongly suggested that reproduc- tive activities in the male are subject to seasonal fluctuations; that is, in fully adult males over 2.5 or 3 years of age, the spermatogenic activity became lower from December to March of every year. It was also clarıfıed in the previous studies that the first appearance of spermatozoa in the seminiferous tubules occurs within 6 or 7 months after birth. The main purpose of this study is to demonstrate the presence of an annual rhythm ın the testicular endocrine function in correlation with the spermatogenic activity as well as the secretory function of fructose from the seminal vesicle, which is a reliable indicator of testicular endocrine function. Another purpose is to determine exactly the completion of sexual maturation in males. ' Presented in part at the International Symposium on Capricornis and Its Related Species, Komono, Mie Prefecture, Japan, May 1986. - ”Supported by Grants-in Aid for Co-operative Research (58362001 and 61560338) from the Ministry of Education, Science and Culture, Japan. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5303-0178 $ 02.50/0 Seasonalıty of reproduction in male Japanese serows 117 Materials and methods Examination items and the number of anımals used are shown in Table 1. From 1982 to 1985, materials for these examinations were collected in Gifu Prefecture, which is centrally located in Japan, during the season authorized to capture the animals; namely from December to March. Animals shot in their habitats were transported to Gifu University for examination, which is situated at 50 to 150 km distance from the sites of capture. The anımals had been dead usually for a few days, Table 1. Examination items and number of anımals Transported materials Fresh materials Examination Dec. 1982 Dec. 1983 Dec. 1984 Dec. 1982 Dec. 1983 items untill untill untill untill untill Mar. 1983 Mar. 1984 Mar. 1985 Mar. 1983 Mar. 1984 Testis sıze 150 170 181 Seminiferous tubule diameter il Population of primary spermatocytes Testosterone concentration in testes Testosterone concentration in serum Androstenedione concentra- tion in testes Androstenedione concentra- tion in serum Fructose concentration in seminal vesicle Body weight Actual number of anımals examined sometimes for a week or more. Most materials were obtained exclusively from the bodies within 7 days after death. Some fresh materials were acquired at the site of capture within a few hours after the anımal’s death. Classification of age groups. The age estimation that underlies this study was based on the tooth eruption-wear patterns (SUGIMURA et al. 1981). The anımals were classified into 6 age groups (Tab. 2). Anımals of groups O0 to 2-]° are immature ones, in which the second dentition ıs not completed, and those of groups 2-I-II to 2-IV-V are adults in which all teeth are permanent. Size of testis. The product of three dımen- sıons of each testis was obtained, and the pro- ducts for paired testes were summed. Table 2. Classification of age groups in Japanese serows Spermatogenic activity. The seminiferous Age group Age in months or years tubule diameter was measured in one to three „a anımals per month for every group; thus, ın 15 2osmonchs youngs and 16 adults. The mean diameter was 10>2months calculated from measurements of 10 cross sec- Ss2. months tions of seminiferous tubules for each individual. Further, the number of primary spermatocytes at e = ‚6 + 1.8 years the pachytene stage per cross section of semi- 4 + 2.7 years! niferous tubule was counted using the fresh tes- Ä ticular materials removed from 13 anımals (one # ; + 3.9 years to three anımals per month in every age group except for 2-IV-V) at the sites of capture within a ! Calculated from the data on the patterns of few hours after the anımal’s death. The mean cementum annulation in the teeth, which number was calculated from measurements of 10 were presented by the courtesy of Dr. S. cross sections for each individual. Miura Testosterone and androstenedione in testicu- 180 T. Tıba, M. Sato, T. Hirano, I. Kıta, M. Sugimura and Y. Suzuki lar tissue. The testes without any pathological changes were used in this assay. Concentrations of testosterone and androstenedione were measured by radioimmunoassay using procedures similar to those described by SHopono et al. (1975) for the determination of estradiol and progesterone in the plasma. One gram of testicular tissue was homogenized and extracted in four steps with two volumes of ether. The extract was evaporated to dryness ın N, gas stream, re-dissolved in 1 ml of ethanol and stored at -25 °C. Before use, 0.1 ml of the stored extract was evaporated and re-dissolved in BSA-PBS to 1:300 and 1:3 for testosterone and androstenedione assay, respectively. [1, 2, 6, 7, 16, 17-H]-testosterone (138 Ci/mmol) and [1, 2, 6, 7 ?H]-androstenedione (85 Ci/ mmol) were used as competitors. Each of them was added to each test assay tube as 0.1 ml solution (8,000 cpm) ın BSA-PBS. The anti-testosterone-11«-succinate-BSA serum and the anti-andro- stenedione-3-Oxıme-BSA serum each were diluted wıth NRS-EDTA-PBS to 1: 6,000. Cross reactiv- ity of the anti-testosterone serum with androstenedione was 2.24%, and that of the anti-andro- stenedione with testosterone 5.43 %. One-fifth ml of testicular extract, 0.2 ml of antiserum and 0.1 ml of tritiated steroid were mixed. After incubation overnight at 4 °C, the steroids, unlabeled and labeled with tritium, were reacted with dextran-charcoal. Following centrifugation at 2,500 g for 15 min, 0.8 ml of the supernatant was removed. Eleven ml of toluene scintillator containing 0.4 % PPO and 0.01 % POPOP was added, and the radioactivity was measured with a liquid scintillation spectrometer. Unknown levels of steroids were read from a standard curve and expressed as nanograms per gram of testicular tissue. In order to investigate the possibility of a lower testicular testosterone level during the period between the anımal’s death and tissue sampling, a comparison was made between the materials obtained, respectively, one to two days and 4 to 5 days after the anımal’s death. There was no significant difference between the former (444.19 # 96.14 ng/g; n=8, adults, Dec.) and the latter (455.27 # 60.01 ng/g; n=8; adults, Dec.). Testosterone and androstenedione in serum. Serum samples were obtained from the coagulated blood in the heart and stored frozen at -25 °C. Concentrations of the steroids were measured with the same method as described above. After thawing the frozen samples, 0.6 ml of serum was extracted in three steps with four volumes of ether. The extract was evaporated to dryness in N, gas stream and re- dissolved in 0.6 ml ethanol. Before use, 0.05 ml of the extract was decanted into another test for testosterone assay, evaporated to dryness, and diluted with 0.2 ml BSA-PBS. For androstenedione assay, 0.2 ml of the extract was treated similarly. Further procedures were the same as described for the steroids in testicular tissue. Concentrations of the steroids were expressed as nanograms per milliliter serum. In order to examine the possibility of a lower testosterone level during the period between the anımal’s death and serum sampling, a comparison was made between the transported and the fresh materials (see Tab. 1). There was no significant difference between the former (3.17 # 0.26. ng/g; n=5, adults, Jan.-Mar.) and the latter (2.94 # 0.44 ng/g, n=5; adults, Jan.-Mar.). Fructose concentration in seminal vesicle: After obtaining the wet weight, the seminal vesicles were stored frozen at -25 °C. But owing to the tiny size of this organ ın young anımals, it was extremely difficult to acquire the minimum volume of material necessary for this chemical assay. Therefore, the seminal vesicles from anımals of age groups 0 and 1 were pooled so as to provide enough material. After thawing of the frozen materials, the fructose contents were measured by the method of LinDnEr and Mann (1960). To deal with the possibility of a lower fructose content during the period from the anımal’s death and seminal vesicle sampling, a comparison was made between the materials obtained, respectively, one and 4 days after the anımal’s death. There was no significant difference between the former (1.27 # 0.43 mg/g; n=5; adults, Dec.) and the latter (1.04 # 0.15 mg/g; n=5; adults, Dec.). Apart from this assay, another experiment was carried out on the influence of storage upon fructose content using a bull’s seminal vesicle. It was proved that the concentration remained unchanged at 4 °C for 7 days. Body weight. As a supplementary method for determining the end of sexual maturation, the body weight was obtained ın 545 anımals for three consecutive seasons. Results and statistical analysis. The mean values obtained have been shown with their standard error. Statistical comparisons between the mean values were made using Student’s t test. Results Size of testis The means and standard errors shown in Fig. 1 were calculated from measurement covering three consecutive capturing seasons. But before these mean values were obtained, the significance of difference in mean value between each single season was examined. No Seasonality of reproduction in male Japanese serows 181 cm?’ M#+SE from > all adults = on 120 (72) 2 36 22 77 56 ! 28 (eP} > es 37 100 h 15 2] 19 15 12 7 3] I} | aa 14 . 80 | 10 7 R H 22 10 Per] 40 20 0 DESNEIM DIFM DJ FM DJIFM DUIFM DJ FM DJ FM Calendar month 0 1 2-1° 2-I-Il 2-Ill 2-IV-V 2-I-II—IV-V Age group Fıg. 1. Seasonal changes in combined size of paired testes for each age group of Japanese serows (M#+SE, Dec. 1982-Mar. 1985). D: December; J: January; F: February; M: March. Numbers above bars represent sample size. In the following the same abbreviations are used significant difference was found among the three seasons, and there was also no significant difference ın size between paired testes. As seen clearly in the figure, the testes grow rapidly between the age groups 0 and 2-I°, that is, from 7 months to about 2.5 years of age. The testes continue to grow slowly in the age groups 2- I-II, and thereafter the testis sıze ıs on a relatively constant level. In adults, a decrease in size occurs every year from December to March. Mean values taken from all the adults belonging to groups 2-I-II to 2-IV-V show a highly significant difference between two consecutive months (p< 0.01). Testosterone concentration in testicular tissue The first appearance ot testicular testosterone is demonstrable in fawns at 7 to 10 months of age (Fig. 2). The concentration rapidly increases in age groups O0 to 2-I-II. In adults, the concentration varıes widely. It is noticeable that, in age group 2-I-II, a significant increase is recognized between January and March (+196.9 %, p< 0.05%). A tendency to increase in the same month is found more or less in other groups except for age group 2-IV-V. But the mean value obtained from all the adults for December is significantly higher than in any other month (p< 0.01), and the increase from January to March is not significant. Testosterone concentration in serum An increase with age in the immature group is uncertain, but seasonal changes in the adults are relatively clear (Fig. 3). The mean value from all the adults in December is significantly higher than in any other months (p< 0.01). No significant difference was found among January, February and March. It was also statistically clarıfıed that there ıs a highly significant correlation between the concentration in serum and that in testicular tissue (r=0.814, p<0.01, n=112). 182 T. Tiba, M. Sato, T. Hirano, I. Kita, M. Sugimura and Y. Suzuki M#+SE from all adults 700 600 500 Testosterone concentration 400 300 200 DI FM DEISESM D FM DIFM DUFM DJ FM DJ FM I J 0 1 2-1° 2-1-11 2-11I 2-IV-V 2-I-II—IV-V Fig. 2. Seasonal changes in testosterone concentration in testicular tissue for each age group of Japanese serows (M+SE, Dec. 1983-Mar. 1985) ng/m] M=SE from all adults 12 zZ [o} = S FR) os 10 © = [e) © eb} e 2 S 8 un = [72] [eb] und 6 ji 2 DIUMEZM DJFM DJ 0 1 2-1° 2-1-II 2-I1I 2-IV-V 2-I-II—IV-V FM DUFM DROEREM DJFM DEJEZEIM Fig. 3. Seasonal changes in testosterone concentration in serum for each age group of Japanese serows (M+SE, Dec. 1983-Mar. 1985) Seasonality of reproduction in male Japanese serows 183 Correlation between spermatogenic activity and testosterone level in testicular tissue With the same anımals in which both the seminiferous tubule diameter and the testosterone concentration in testicular tissue were measured, a correlation between the two was statistically analyzed. A highly significant correlation-coefficient was found (Fig. 4). On the other hand, in the fresh testicular materials obtained shortly after the animal’s death, a highly significant correlation was found between the mean diameter of the seminiferous tubule and the mean number of the primary spermatocytes at the pachytene stage per cross section of the tubule (Fig. 5). ng/g ° 5 800 D) = r=0.532** (P(0.01) er = n=3] S 600 un 30 = A 8 r=0.820** (P£0.01) s ° o wo e 3 n=13 S 400 5 = 20 . = oo. En o 2 ; % - : Ri s > 5 [7] ® .e s 1%} o 4= 7 .° ‚w 200 . l 3.10 io ® ° ® SOG = oo o >) 0 ROSEN. = 40 DE JR 130 150 170 190 210 230m 120 140 160 180 200 220m Diameter Diameter Fıg. 4 (left). Correlation between seminiferous tubule diameter and testosterone concentration in testicular tissue of Japanese serows. — Fig. 5 (right). Correlation between seminiferous tubule diameter and number of pachytene primary spermatocytes per cross section of seminiferous tubule in Japanese serows Androstenedione concentration in testicular tissue In more than half of all anımals examined (66/126), the androstenedione levels were under the lowest detection limit of the assay (45 pg/g). For this reason, no mean values were calculated, and the measured values were plotted for each individual in Fig. 6. Androstenedione concentration in serum As shown in Fig. 7, it ıs very difficult to find a decided tendency in the fluctuation of mean values for each age group. In the mean values calculated from all adults, the value for December is highest, showing a highly significant difference from that for February (p<0.01). The second highest value in March shows a significant increase over February (p< 0.05), but is not significantly different from that in December. Fructose concentration in seminal vesicle The concentration of fructose per gram of seminal vesicle was determined (Fig. 8). For the reasons which have been advanced, however, no sufficient data for statistical analyses were obtained in the immature group, but the values obtained from young animals seem to be on a lower level than in adults. A decrease from December to March is statistically significant (p<0.01) in the mean values calculated from all adults. A statistical analysis of the correlation between fructose concentration in seminal vesicle and testosterone concentration in serum was conducted with the same animals in which both substances were evaluated. A highly significant correlation-coefficient was found between the two (r=0.75; p<0.001; n=38). 184 T. Tıba, M. Sato, T. Hirano, I. Kıta, M. Sugimura and Y. Suzuki Androstenedione concentration DJIFM DIJFM DAN DJ IEEM DEDRERM DEREM 0 1 2=1° 2-1-11 2-Ill 2-IV-V Fig. 6. Seasonal changes in androstenedione concentration in testicular tissue for each age group of Japanese serows. Spots on the abscissa indicate values below the lowest limit (45 pg/g) ng/m] M£SE from all adults 0.8 Androstenedione concentration m 0 1 DJ FM DIFEM DJ 2 FM DRUNTERM DJUFM DJ REM DJ FM I 2-T-I1 2-111 2-IV-V 2-I-II—IV-V Fig. 7. Seasonal changes in androstenedione concentration in serum for each age group of Japanese serows (M+SE) Seasonality of reproduction in male Japanese serows 185 Body weight It is clear from Fig. 9 that the growth of male animals ends at about 2.5 or 3 years of age, and thereafter the body weight is maintained on a relatively constant level. Ata glance, the body weight appears to decrease steadily from December to March. However, the difference between the highest and the lowest values in each age group is not everywhere significant; thus, it ıs significant in group 1 (December to February: -19%; p< 0.01), group 2-I-II (December to March: -9.4%; p<0.01) and group 2-III (December to February: -9.9%; p=0.01), but not in the other groups. mg/g M+SE from all adults Fructose concentration DEIBIESM DRJREEIM DIFM DRIBERM DEIREEM DRJEREIM DJ FM 0 1 2-I° 21T 2-Ill 2-IV-V 2-1-II—IV-V Fig. 8. Seasonal changes in fructose concentration in seminal vesicle for each age group of Japanese serows (M+SE, Dec. 1982-Mar. 1985). The values for age groups O0 and 1 were obtained from pooled material, except for the mean value for age group I ın December (see text) Body weight DJ FM DNERM DJFM DIEEEM DJFM DJ FM 0 1 2-1° ZSCORT 2-IIl 2-IV-V Fıg. 9. Seasonal changes in body weight for each age group of Japanese serows (M+SF, Dec. 1982-Mar. 1985) 186 T. Tıba, M. Sato, T. Hirano, 1. Kita, M. Sugimura and Y. Suzuki Discussion From experiments ın vitro about the metabolic routes of steroid hormone biosynthesis in the Japanese serow’s testis (NAKAMURA and Suzukı 1985), it has been clarified that the ultimate product of androgen synthesis is testosterone in this species, too. And the major metabolic route in the testicular formation of testosterone from pregnenolone is A>- pathway. It has also been demonstrated that androstenedione is produced only at a very low rate via the other minor metabolic route, A*-pathway. These facts coincide with our own findings that androstenedione levels in testicular tissue as well as in serum are very low. It, therefore, can hardly be ımagined that androstenedione plays an important role in the male reproductive functions of this anımal. This supposition seems to be supported by our results that ıt was almost impossibile to find a relationship between the fluctuation in adrostenedione levels and advancıng age or season. In the goat, the domestic ruminant related most closely to the Japanese serow, the end- products ın testicular formation of androgens from pregnenolone are testosterone and 17«, 20«-dihydroxy-4-pregnen-3-one; and testosterone is synthesized through both A*- and A°-pathways (Mort et al. 1980). As for the androstenedione levels in domestic ruminants, it is well known ın the bull that the ratio androstenedione/testosterone decreases with age (LINDNER and Mann 1960). Concerning changes in adrostenedione levels with advancing age ın the young goat, there ıs a discrepancy in the literature. LEiDr et al. (1970) reported, in measuring testicular androgens that there was no change in the ratio androstenedione/ testosterone. On the contrary, SAUMANDE and ROUGER (1972) mentioned from the result of determination of the plasma testosterone level in one young goat of Saanen breed, that during the period of lower testicular activity androstenedione is more important than testosterone. In the ram, the testicular concentration of testosterone is always higher than that of androstenedione between the 5th and 12th month of life (Eıx-Nes 1975). The sexual maturation in this anımal occurs just in this period (HULET and SHELTON 1980). Our previous indication that the testicular functions in Japanese serows are subject to seasonal fluctuation (TıBA et al. 1981a), is now clearly demonstrated by the findings that the testosterone levels in the testicular tissue as well as in serum are significantly higher ın December than in any other months. It has been also clearly demonstrated in this study that the fructose content in the seminal vesicle is seasonally changeable, depending upon the testosterone level. However, we do not yet know all about the seasonal fluctuations in this anımal’s reproduction. The height of the breeding season is, as mentioned before, October and November, when the concentrations of steroids and fructose must be placed on even higher levels than in December. On the contrary, the spermatogenic activity in the goat does not fluctuate ın parallel with the testosterone levels ın testicular tissue (LEipL et al. 1970). According to these authors, “The seasonal rhythm of reproduction in the male goat affects only the functional relationship “androgens-accessory sexual glands-seminal plasma and contents’, whereas the germinal cells show no fluctuations.” And the authors add: “These findings suggest that spermatogenesis continues to be stimulated by amounts of hormones which are insufficient to stimulate the accessory sexual glands to their full function.” It ıs clear from the results that the sexual maturation in this anımal is completed at 2.5 to 3 years of age. This coincides with the report mentioned earlıer that the first parturition in female occurred usually at 3 years of age. In our own studies on morphology of ovaries and fetuses, female serows become sexually mature at about 2.5 years of age. The youngest pregnant females were 30 months old (Kıra et al. 1983). If male Japanese serows reach puberty within 6 or 7 months after birth and come to full sexual maturation at 2.5 to 3 years of age, there is an interval of 2 to 2.5 years between the beginning and the end of sexual maturation. This interval appears extremely long as compared with domestic Seasonality of reproduction in male Japanese serows 187 ruminants. Bulls (Hawk and BELLows 1980), rams and goats (HuULET and SHELTON 1980) are brought into service either via natural or artificial insemination after an interval of a few to 6 months, or at most one year following puberty. Such a comparison, however, may not be reasonable, because in domestic animals the sexual maturation is hastened bya high level of nutrition, artificial selection and other beneficial conditions. In wild ruminants, we were not able to find any reports in which the beginning and the end of sexual maturation are separately described. Acknowledgement The authors wish to thank Dr. T. NAKAMURA for providing informations on RIAs. Zusammenfassung Jahresrhythmische Veränderungen von Fortpflanzungstätigkeiten und Geschlechtsreife beim männlichen Japanischen Serau (Capricornis crispus) Die Fortpflanzungstätigkeit wurde bei zahlreichen männlichen Exemplaren des Japanischen Serau untersucht, die 1982-1985, von Dezember bis März, in der Provinz Gifu in Zentraljapan erlegt wurden. Die Prüfungen erstreckten sich auf spermatogenetische Tätigkeit, Testosteron- und Andro- stendionkonzentration im Hodengewebe sowie im Serum und auf Fruktosekonzentration in der Samenblase. Bei vollwüchsigen Männchen über 2,5 bis 3 Jahre wurde der Maximalwert bei jeder Prüfung fast ausnahmslos im Dezember gewonnen. Von Dezember bis ın den März hinein, zeigte sich eine offensichtliche Tendenz zur Herabsetzung der Werte. Eine hochsignifikante Korrelation wurde festgestellt, sowohl zwischen der spermatogenetischen Tätigkeit und der Testosteronkonzentration, als auch zwischen der Fruktosekonzentration und dem Testosteronspiegel. Daraus geht schon hervor, daß die männlichen Fortpflanzungstätigkeiten dieser Spezies den jahreszeitlichen Schwankungen unterworfen sind. Bei jungen Tieren steigerten sich die spermatogenetische Aktivität und die Testosteron- sowie Fruktosekonzentration mit zunehmendem Alter und erlangten das Niveau des Erwachsenen im Alter von 2,5—3 Jahren. Die Androstendionkonzentration im Serum bei Jungen steht auf gleicher Höhe mit der von Erwachsenen, und es gab keine deutliche Tendenz zur Herabsetzung der Werte im Laufe der sexuellen Reifung. Literature ARASARA, T. (1979): Social organization of Japanese serow in Nibetsu, Akıta Pref. especially on the social units. Rep. Nat. Conserv. Soc. Japan 56, 87-102. ARKASAKA, T.; MaruYamA, N. (1977): Social organization and habitat use of Japanese serow in Kasabori. Jap. J. Mammal. Soc. Japan 7, 87-102. EıK-Nes, K. B. (1975): Biosynthesis and secretion of testicular steroids. In: Handbook of Physiology. Sect. 7: Endocrinology 5, Male Reproductive System. Ed. by D. W. Hamilton and R. ©. Greep, Washington: Amer. Physiol. Soc. pp. 95-115. Hawk, H. W.; BErrows, R. A. (1980): Reproductive cycles: Beef and dairy cattle. In: Reproduction in farm anımals. Ed. by E. S. E. Hafez. Philadelphia: Lea and Febiger, pp. 337-345. HULET, C. V.; SHELTON, M. (1980): Reproductive cycles: Sheep and goats. In: Reproduction in farm anımals. Ed. by E. S. E. Hafez. Philadelphia: Lea and Febiger. 347-357. Iro, T. (1971): On the oestrous cycle and gestation period of the Japanese serow, Capricornis crispus. Jap. J. Mammal. Soc. Japan 104-108. (In Japanese wıth English summary). Kıra, I.; SUGIMURA, M.; Suzukı, Y.; TıBa, T. (1983): Reproduction of wild Japanese serows based on the morphology of ovaries and fetuses. Proc. Vth. World Conf. Anım. Product., Tokyo, 2 243-244. Komorı, A. (1975): Survey on the breeding of Japanese serows Capricornis crispus, in captivity. J. Jap. Ass. Zool. Gardens and Aquariums 18, 53-61. (In Japanese with English summary). LEıpr, W.; HorFmann, B.; Karc, H. (1970): Endokrine Regulation und jahreszeitlicher Rhythmus der Fortpflanzung beim Ziegenbock. Zbl. Vet. Med. A. 17, 623-633. LINDNER, H. R.; Mann, T. (1960): Relationship between the content of androgenic steroids in the testes and secretory activity of the seminal vesicles in the bull. J. Endocrinol. 21, 341-361. Morı, M.; MATSUKURA, S.; KAwWAKURA, K.; Tamaokı, B. (1980): In-vitro synthesis of androgen from pregnenolone in the testes of the goat (Capra hircus) and identification of 5-pregnene-3ß,17«, 20«@-triol as an intermediate in the metabolic pathway of pregnenolone. J. Endocrinol. 84, 381-390. NAKAMURA, T.; Suzukı, Y. (1985): Studies on the steroid hormone biosynthesis in the wild Japanese serow’s testes. Basic studies on reproduction, morphology, diseases and population trends in Tıba, M. Sato, T. Hirano, 1. Kıta, M. Sugimura and Y. Suzuki 188 Japanese serows 1985. Grant-in-aid for co-operative research report (58362001) 107-116. (In Japanese). SAUMANDE, J.; ROUGER, Y. (1972): Variations saisonnieres des taux d’androgenes dans le plasma de sang p&@ripherique chez Bouc. C. R. Acad. Sc., Paris 274, 89-92. SHODONO, M.; NAKAMURA, T.; TANABE, Y.; WAKABAYASHI, K. (1975): Simultaneous determinations of progesterone and luteinizing hormone in the plasma during the ovulatory cycle of the hen. Acta endocrinol. 78, 565-573. - SUGIMURA, M.; SUZURI, Y.; KAMIya, $.; FuJITa, T. (1981: Reproduction and prenatal growth in the wild Japanese serow, Capricornis crispus. Jpn J. Vet. Scı. 43, 553-555. TıBA, T.; SuGımurA, M.; Suzukı, Y. (1981a): Kinetik der Spermatogenese bei der Wollhaargemse (Capricornis crispus) I. Geschlechtsreife und jahreszeitliche Schwankung. Zool. Anz. 207, 12-24. —— (1981b): Kinetik der Spermatogenese bei der Wollhaargemse (Capricornis crispus) 11. Samenepithelzyklus und Samenepithelwelle. Zool. Anz. 207, 24-34. Authors’ addresses: T. TıBa, Department of Theriogenology, Faculty of Agriculture, Gifu Univer- sity, Gifu 501-11, Japan; M. Saro, Handa Health Centre of Aichi Prefecture, Handa, Aichi 475, Japan; T. Hırano, Institute of Safety Research, Ono Phar- maceutical Company Limited, Mikuni, Fukui 913, Japan; I. KıTa, Department of Theriogenology, Faculty of Agricultue, Gifu University, Gifu 501-11, Japan; M. SuGımURA and Y. Suzukı, Department of Veterinary Anatomy, Faculty of Agriculture, Gifu University, Gifu 501-11, Japan Z. Säugetierkunde 53 (1988) 189-190 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 NZISSENSCHIAIETIELEHEIEZKTURZMTIETETEUNG An unusual record of Hooded seal (Cystophora cristata) in SW Spain By C. IBANEZ, M. DELIBES, J. CASTROVIEJO, RoSALIA MARTIN, J. F. BELTRAN and S. MORENO Estaciöon Biologica de Donana, CSIC, Sevilla, Spain Receipt of Ms. 3. 8. 1987 The Hooded seal, Cystophora cristata, ıs a typical Arctic species from the central and western areas of the northern Atlantic. Its distribution extend from Newfoundland and Labrador to Greenland, Iceland, Bear Island and Spitsbergen (Reeves and Ling 1981; Kovacs and LAVIGNE 1986). In very rare occasıons, wandering solitary individuals have been reported on both sides of the Atlantic as far south as Florida (USA) and Portugal (Kınc 1983). On February 26, 1983, a specimen of Hooded seal was found dead on the beach near Torre Zalabar in Donana National Park, 10 km NW of the mouth of the Guadalquivir river ın the Province of Huelva, Spain (36° 54'N, 6° 24'’W). The specimen was a pregnant adult female and showed a deep wound, measuring approximately 20 cm, on its right side. The death was recent as the anımal was well preserved except for the eyes which had been eaten. The skin, the skeleton and the foetus are deposited in the collection of the Estacıiön Biolögica de Donana (CSIC, Sevilla). External and cranıal measurements (in millimeters), the latter taken according to WrIG and LıE (1984), and the weight (in kilograms) of the specimen were as follows: total length (nose-tail), 2010; tail length (from anus), 155; posterior extremity length (with cartılage), 350; anterior extremity length, 303; condylobasal length, 2411; basal length, 225; palatal length, 134; upper tooth-row length, 78; zygomatic width, 185; auditory meatus width, 157; palatal width, 60; orbital width, 33; mandible length, 158; weight (with foetus), 116. Except for the weight, which was greatly inferior to that considered normal for an adult female (Reeves and Ling 1981; Kovacs and LAvIGnE 1986), these measurements are among the highest reported for the species, corresponding to an age well over 20 years (WrıG 1985). The total length of the foetus was 75 and the weight 9.5. The nearest localıties to Donana where the species has been reported are two ın France (I. Oleron, in Charente-Maritime, and Bidasoa) (POUVREAU et al. 1980) and three ın Portugal (Buarcos, Algarve and Peniche) (REINER 1979, 1980). All these cases dealt with young individuals, observed between June and August. Although the reports from Bidasoa and Algarve are close to the Spanish coast, this is the first concrete record for Spain and the furthest from the habitual range of the species. The importance of this new record, however, lies in the fact that it deals with an adult individual, as opposed to the others in these latitudes which were undoubtedly related to juvenile dispersion. Although of advanced age, this specimen cannot be considered senile, as it presented normal reproductive activity. At the time of death, it was in poor physical condition. The subcutaneous layer of fat, which reaches a tickness of 4.8 cm at this time of year (REEvES and Ling 1981), was reduced to 1 cm and the specimens’s weight was, as previously mentioned, well below normal. The evidence would indicate the seal died after depleting U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5303-0189 $ 02.50/0 190 C. Ibanez, M. Delibes, J. Castroviejo, Rosalia Martin, J. F. Beltran its reserves, perhaps as a result of the wound in its side which made it incapable of obtaining food. The decrease in its faculties probably led to its being swept by the ocean streams to the Spanish coast. References Kıng, J. E. (1983): Seals of the world. London: British Museum (Nat. Hist.). Kovacs, K. M.; LavicnE, D. M. (1986): Cystophora cristata. Mammalıan Species 258, 1-9. Pouvreau, B.; Ducvy, R.; Arızev, C.; Basın, P. (1980): Capture d’ un phoque & crete, Cystophora cristata (Erxleben, 1777) sur la cöte frangaise atlantique et recherches sur sa pathologie. Bull. Cent. Etude. Rech. scı. Biarritz 13, 7-12. Reevss, R. R.; Ling, J. K. (1981): Hooded seal. Cystophora cristata Erxleben, 1777. In: Handbook of marine mammals. Vol. 2: Seals. Ed. by S. H. Rınaway and R. J. Harrıson. London: Academic Press. REINER, F. (1979): Notas sobre a ocorrencıa de alguns pinipedes na costa Portuguesa. Mems. Mus. Mar, Portugal, 1, 1-24. — (1980): Nota sobre a segunda ocorrencia de uma foca de mitra, Cystophora cristata (Erxleben, 1777) nas costas de Portugal. Mems. Mus. Mar, Portugal, 1, 1-6. Wis, ©. (1985): Morphomertric varıatıon in the Hooded seal (Cystophora cristata). J. Zool. Lond. (A) 206, 497-508. Wırc, ©.; Lie, R. W. (1984): An analysis of the morphological relationships between the Hooded seals (Cystophora cristata) of Newfoundland, the Denmark Strait, and Jan Mayen. J. Zool. Lond. (A) 203, 227-240. Authors’ address: CARLOS IBANEZ, MIGUEL DELIBES, JAVIER CASTROVIEJO, ROSALfA MARTfN, JUAN FRANcISCO BELTRAN and SACRAMENTO MORENO, Estaciön Biolögica de Donana, Apartado 1056, E-41080 Sevilla, Spain BUCHBESPRECSEUNGEN NauTa, W.]J. H.; FEırTAG, M.: Fundamental neuroanatomy. New York: Freeman and E03 1986.34 pp 116 figs. £ 18.95. ISBN 02107172323 In diesem Buch geben die bekannten Autoren einen Überblick über den Grundbauplan des Säugetier- gehirns, vornehmlich am Beispiel des Menschen. Sie haben die Stoffülle in drei große Abschnitte unterteilt und gehen in üblicher Weise synthetisch vor. Im einleitenden Kapitel wird sehr kurz auf frühe stammesgeschichtliche Entstehung nervöser Strukturen und auf die Situation bei basalen Evertebraten eingegangen. Es folgt eine Beschreibung von Neuronen und Gliazellen in Bau und Funktion (Darstellungsmethoden, Möglichkeiten von Neuronenkontakten, Mechanismen der Reiz- leitung, Transmittersubstanzen etc.). In einer Übersicht werden dann die großen Hirnteile in ihrer Ontogenese und adulten Organisation skizziert sowie Prinzipien retrograder und anterograder Degenerationstechnik. Fin zweiter Abschnitt ist den Verbindungen und Funktionssystemen im Gehirn gewidmet, indem auf die großen afferenten und efferenten Systeme, auf neocorticale Funktio- nen, viscerale Innervation und limbisches System eingegangen wird. Der dritte Abschnitt schildert die anatomischen Verhältnisse. An Hand von Querschnittspräparaten und Zeichnungen werden in Form eines Hirnatlas Kerngebiete und Faserzüge der verschiedenen Hirnteile des Menschen benannt und besprochen. Das Buch ist in klarer Sprache abgefaßt und verständlich geschrieben. Die wichtigsten Fortschritte in der Geschichte der Hirnforschung sowie Erklärungen und Ableitungen sehr vieler wissenschaftli- cher Namen sind geschickt in den Text eingeflochten. Die Bebilderung mit Originalpräparaten von außerordentlicher Qualität, Blockdiagrammen und Schaltschemata beeindruckt und gewinnt zusätz- lich an Wert durch besonders ausführliche, erläuternde Unterschriften. Insgesamt liegt eine gelungene und kurzgefaßte Informationsquelle über dieses Organ vor, die jeden Interessenten ansprechen muß. D. Kruska, Kiel HiILDEBRAND, M.; BRAMBLE, D. M.; Lıem, K. E.; WAaAxE, D. B. (eds.): Functional Vertebrate Morphology. Cambridge Mass., London: The Belknap Press ot Harvard University Press 1985. 430 pp. US $ 40.25. ISBN 0-674-32775-6 Die langwährende Stagnation der vergleichenden Morphologie ist heute weitgehend überwunden. Sie beruhte letzten Endes auf der Reduktion der Formenkunde auf einen geschlossenen, typologischen Formbegriff. Die Einsicht in den essentiellen Zusammenhang und die Untrennbarkeit von Form und Funktion in der Organismenwelt führte in den vergangenen zwei Jahrzehnten zur Überwindung der Erstarrung und zu neuen Denkweisen. Diese sind gekennzeichnet durch die Einbeziehung experimen- teller und quantitativer Methoden, durch Berücksichtigung von Lebensweise, Verhalten und Umwelt, durch Einführung neuer technischer Verfahren (z.B. Bewegungsanalyse durch Elektromyographie und vieles andere). Umwelt und Organismus sind historischen Einflüssen unterworfen. Daher bleibt die Beachtung des zeitlichen Faktors (Evolutionsbiologie), der schon früher eng mit der Morphologie verknüpft war, eine notwendige Voraussetzung. Schließlich vermag die Analyse des individuellen Zyklus (Embryologie) grundsätzliche Einsichten zu vermitteln. Das vorliegende Buch ist ein hervor- ragendes Beispiel für den Wandel in Denk- und Arbeitsweisen der vergleichenden Morphologie heute. Das von 20 Autoren bearbeitete Werk setzt Grundkenntnisse der Anatomie, Physiologie und Formenkunde voraus, ist also nıcht für den Anfänger geeignet. In zwei einleitenden Kapiteln werden funktionelle Anpassung bei Skelettstrukturen (leider ohne Berücksichtigung der Arbeiten von Pau- wELs und Kummer) und Probleme der Körpergröße und Allometrie behandelt. Das Kernstück des Werkes sind die sieben Beiträge über verschiedene Lokomotionsweisen (Laufen, Springen, Klettern, subterrane Lokomotion, Schwimmen, Fliegen, Bewegung ohne Gliedmaßen). Es folgt je ein Beitrag über Energetik der Lokomotion und Atmung. Drei Kapitel befassen sich mit Biologie der Nahrungs- aufnahme - Kauakt. Zwei Beiträge sind den Sinnessystemen (Octavo-Lateralis und Auge) gewidmet. Die zentralnervöse Kontrolle und Koordination wird in einem inhaltsreichen Kapitel berücksichtigt. WARE und LiEMm fassen in einem Schlußbeitrag „Morphology, current approaches and concepts“ die derzeitige Situation in einer Synthese zusammen. Skelett und Lokomotion stehen ganz ım Vorder- grund der Darstellung (nahezu die Hälfte des Inhaltes). Ursache hierfür ist natürlich die Tatsache, daß Skelett und Lokomotion am leichtesten der morphologischen und physikalisch-technischen Untersu- chung zugänglich sind. Die Ausstattung des Buches ist hervorragend. Hervorgehoben sei, daß die verschiedenen Beiträge inhaltlich und stilistisch außerordentlich gut aufeinander abgestimmt sind und sich so - im Gegensatz zu manchen Büchern mit vielen Autoren - ein geschlossenes Ganzes ergibt. Ihm ist weite Verbreitung zu wünschen, zumal es neue Denkweisen und Methoden in einer verjüngten Wissenschaft dokumen- tiert. D. STARcK, Frankfurt/M. 192 Buchbesprechungen Kress, J. R.; Davies, N. B: An Introduction to Behavioural Ecology. 2. Ed. Oxford: Blackwell Scientific Publications 1986. 400 pp., 121 ıll., £ 12.80. ISBN 0-632-01498-9 cc Die erste Auflage der „Einführung...“ in dieses moderne Gebiet der Ethologie war sowohl in der englischen als auch in der deutschen Ausgabe ein Erfolg. Es ist die augenblicklich einzige auf dem Markt zu erhaltende Einführung in die Verhaltensökologie. So ist es zu begrüßen, daß eine 2. Aufl., die um ca. 100 S. erweitert und an zahlreichen Stellen ergänzt wurde, erschienen ist. Fast alle Kapitel wurden wenigstens zum Teil neu verfaßt, und es sind neue Schwerpunkte hinzugekommen, wie z.B. die Kapitel über „Konkurrenz und Ressourcen“ und über „Selbstsucht und Altruismus“. Erweitert wurden die Kapitel „Kämpfen und Einschätzen“ sowie über „elterliche Fürsorge und Paarungssy- steme“. Das Literaturverzeichnis, das um 8 S. verlängert wurde, spiegelt den Stand der augenblickli- chen einschlägigen Literatur gelungen wider. R. SCHRÖPFER, Osnabrück SAVAGE, R. J. G.; Lonc, M. R.: Mammal evolution: an illustrated guide. London: British Museum (Nat. Hist.) 1986. 259 pp. £ 17.50. ISBN 0-565-00942-7 In this book, the British palaeontologist R. J. G. SAvAGE gives a short overall description of the modern knowledge on mammalian phylogeny. The subject-matter is presented in 13 chapters starting with general informations on the geology of fossılisation and natural preservation processes, continen- tal drift during earlier times, dating methods and geological times. Then, a characterisation of mammals and a higher classification of this anımal group is given. Furtheron, bones and teeth as well as other anatomical peculiarities of different adaptive forms are outlined and characteristics to differentiate reptiles and mammals. Mammal like reptiles and early mammals are sketched. The descriptions on ancient and recent mammals as well as their distribution, phylogenetical development and evolution are done in nine further chapters headlined as: insectivores, carnıvores, paddlers and swimmers, gliders and fliers, gnawers, early rooters and browsers, mammals on island and continents, hoofed herbivores, and men and monkeys. Consequently, the author preferred a summarized description in connection with special adaptations and lite-style types. In connection with and additional to the text there ıs a consıderable amount of illustrations done by the junior author M.R. Lonc. These illustrations are high quality drawings of species’ habits as well as anatomical details (bones, teeth, skulls, intestines, endocasts, etc.). Very impressive coloured restorations of ancient forms in ancıent landscapes are especially welcome to the reader because of their imaginative value. The book can be recommended not only to mammalogists but also to persons with general interests as a source of overall information. D. Kruska, Kiel SCHEUNERT, A.; TRAUTMANN, A.: Lehrbuch der Veterinär-Physiologie. 7., völlig neube- arb. Aufl., hrsg. von G. WITTEE. Berlin und Hamburg: Paul Parey 1987. 739 S., 418 Abb., 116 Tab. DM 198,-. ISBN 3-489-66216-4 Dieses Lehrbuch der Veterinär-Physiologie nımmt seit der 1. Aufl., von SCHEUNERT, TRAUTMANN, Krzywanek 1939 begründet, einen festen Platz in der Ausbildung der Tiermediziner ein. Ausgewie- sene Wissenschaftler waren auch in der Folge an den Gestaltungen weiterer Auflagen beteiligt, die 6. war bereits 1982 nicht mehr erhältlich. Nun liegt, bedauerlicherweise verzögert, die 7., von G. WITTKE herausgegebene Auflage vor. Die Bearbeitung der verschiedenen Kapitel teilen sich 16 Wissenschaftler in folgender Konzeption: Allgemeine Grundlagen der Lebensvorgänge (EDER/ WITTRE/WELS), Stoff- und Energiewechsel (SCHARRER/PFEFFER/GROPP/MÄNNER/BRONSCH/ZUK- KER), Ihermophysiologie (HÖRNICKE), Blut und Lymphe (Ever), Blutkreislauf (SPÖRRI), Atmung (SPÖRRI/WITTKE), Wasser- und Elektrolythaushalt, Physiologie der Niere (HIERZOLZER/FROMM), Endokrinium (BAMBERG), Fortpflanzung (ZEROBIn), Eibildung (MÄNNER), Nervensystem und Sinnesorgane (WITTKE), Bewegung (WELS), Verhalten (HoFECKER). Gegenüber der vorangegangenen Auflage sind mehrere Abschnitte revidiert, erweitert und modernisiert, andere neu aufgenommen worden. Auf Darstellung biochemischer Prozesse wurde wegen der Ausgliederung dieses heute eigenständigen Faches und entsprechender Ausstattung mit eigenen Lehrbüchern verzichtet. Den Erfordernissen einer einsichtigen Haustierhaltung entsprechend, wurde auch die Verhaltensphysiolo- gie neu bearbeitet. Die Ausführungen sind kurz und prägnant, und im Hinblick auf die zahlreichen Autoren zeigt der Text eine erwähnenswerte, flüssige Einheitlichkeit. Obwohl in erster Linie an klinisch-medizinischer Anwendung orientiert, enthält dieses Lehrbuch sehr viel Information, so daß es auch Säugetierkundlern allgemein empfohlen werden kann. D. Kruska, Kiel Deutsche Gesellschaft für Säugetierkunde 61. Hauptversammlung in Berlin, 27. September bis 1. Oktober 1957 Kurzfassungen der Vorträge und Posterdemonstrationen. Herausgegeben von Prof. Dr. Heinz-Georg Klös, Dr. Hans Frädrich, Prof. Dr. Carsten Niemitz, alle Berlin. Zusammenstellung: Christel Schmidt, Bonn. 1987. 57 Seiten. 24,5x 16,5 cm. Kartoniert 24,- DM Nach 30 Jahren fand die Hauptver- sammlung der Deutschen Gesellschaft für Säugetierkunde wieder in Berlin statt. Die überaus große Zahl der an- gemeldeten Vorträge und Poster be- wies, daß die nunmehr 750jährige Metropole - sie ist Gründungsort der Gesellschaft - für deren Mitglieder und Freunde an Anziehungskraft nichts eingebüßt hat. Säugetierfor- schung wird in Berlin an verschiede- nen Fachbereichen der Freien Univer- sität, der Technischen Universität sowie im Zoologischen Garten betrie- ben; ein Naturkundliches Museum, das nach der Teilung im Westteil der Stadt lange fehlte, befindet sich nun ım Aufbau. Eın Hauptziel der Deutschen Gesell- schaft für Säugetierkunde ist, auf ihren Jahrestagungen über Säugetiere arbei- tende Wissenschaftler verschiedenster Fachrichtungen zusammenzuführen, den Gedanken- und Erfahrungsaus- tausch anzuregen, um so Erkenntnisse aus den einzelnen Forschungsgebieten zu integrieren. Kommunikation, Fort- pflanzung und Funktionsmorphologie der Säugetiere waren 1987 die Schwer- punkte, über die berichtet wurde. Sie bieten durch ihre Vielfalt die Grund- lage für einen intensiven Gedanken- und Erfahrungsaustausch, eines der Hauptziele der Gesellschaft. Der Wunsch der Veranstalter, daß Berlin als Stätte des Dialoges den geeigneten Rahmen bilden würde, hat sıch voll erfüllt. Zu beziehen durch jede Buchhandlun Verlagsbuchhandlung PAJL Paul Parey Spitalerstraße 12 Dee 2000 Hamburg 1 Erscheinungsweise und Bezugspreis 1988: 6 Hefte bilden einen Band. Jahresabonnement Inland: 308,- DM zuzüglich 13,80 DM Versandkosten, Gesamtpreis 321,80 DM einschließlich 7% Mehrwert- steuer. Jahresabonnement Ausland: 308,- DM zuzüglich 18,- DM Versandkosten. Das Abonne- ment wird zum Jahresanfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, Spitalerstraße 12, D-2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesellschaft für Säugetier- kunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. Z. Säugetierkunde 53 (1988) 3, 129-192 Winfried Ahne Grundriß der Zoologie für Tiermediziner Von Prof. Dr. rer. nat., Dr. med. vet. habil. Winfried Ahne, Institut für Zoologie und Hydrobiologie, Tierärztliche Fakultät der Lud- wig-Maximilians-Universität München. 1986. 308 Seiten mit 133 Abbildungen und 11 Tabel- len Kartoniert DM 39,-. Dieser Studientext basiert auf der langjäh- rigen Lehrtätigkeit des Verfassers im Fach Zoologie für die Studierenden der Tiermedi- zin an der Ludwig-Maximilians-Universität in München. Die Stoffauswahl richtet sich weitgehend nach den Bestimmungen der Approbationsordnung für Tierärzte und ent- hält im wesentlichen den in der Studienord- nıng enthaltenen Lehrzielkatalog für das Fach Zoologie. Neben dem systematischen Überblick über die Tierstämme werden ins- bsondere Grundkenntnisse der Genetik, der Fortpflanzung und der Entwicklung, der Zytologie und der Histologie sowie der Evo- lution, der Abstammungslehre und der Ökologie vermittelt. Ausgehend von der rein zoologischen The- matik wird auch das vielseitige Aufgaben- gebiet des Tierarztes berücksichtigt. Dazu werden sowohl die Tiergruppen ausführlich im Kapitel „Spezielle Zoologie“ beschrie- ben, die in der tierärztlichen Tätigkeit eine Rolle spielen, wie etwa die klassischen Haus- tiere, als auch neue, tierartbezogene Betäti- gungsfelder angesprochen, wie z.B. die tier- ärztliche Betreuung von Aquakulturen (Weichtiere, Krebse, Fische) und Bienen- bestände oder die Behandlung von wechsel- warmen Wirbeltieren (Fischen, Amphibien, Reptilien), die spezielle zoologische Kennt- nisse auch über exotische Tierarten voraus- setzen. In erster Linie als Vorlesungsbegleittext und Nachschlagewerk für die Studierenden der Tiermedizin geschaffen, vermittelt das Buch aber auch fächerübergreifende Grundlagen und bietet dem praktizierenden Tierarzt eine schnelle und übersichtliche Orientierungs- hilfe. Aus dem Inhalt: Allgemeine Zoologie Die Zelle - Das Gewebe - Genetik - Fortpflan- zurg - Entwicklung - Evolution - Ökologie - Zoologische Systematik. Spezielle Zoologie - Protozoa (Urtiere, Einzeller) - St. Sarcomasti- gophora - St. Labyrinthomorpha - St. Api- complexa - St. Microspora - St. Ascetospora - St. Myxozoa - St. Ciliophora. Metazoa (Gewebetiere) - St. Mesozoa - St. Schwämme (Porifera) - St. Nesseltiere (Cnidaria) - St. Acnidaria. Bilateria - Protostomia - St. Platt- würmer (Plathelminthes) - St. Schnur- würme (Nemertini) - St. Kelchwürmer (Kampotozoa) - St. Schlauchwürmer (Ne- mathelminthes) - St. Priapswürmer (Priapu- lida) - St. Weichtiere (Mollusca) - St. Spritzwürmer (Sipunculida) - St. Igelwür- mer (Echiurida). Gliedertiere (Articulata) - St. Ringelwürmer (Annelida) - St. Stummel- füßer (Onychophora) - St. Bärtierchen (Tar- digrada) - St. Zungenwürmer (Linguatulida) - St. Gliederfüßer (Arthropoda) - St. Kranz- fühler (Tentaculata). Deuterostomia - St. Pfeilwürmer (Chaetognata) - St. Stachel- häuter (Echinodermata) - St. Kragentiere (Branchiotremata) - St. Bartwürmer (Pogo- nophora) - St. Chordatiere (Chordata) - Anhang - Literaturverzeichnis - Bildquel- lennachweis - Sachverzeichnis. Berlin und Hamburg PAUL PAREY Vol. 53 (4), 193-256, August 1988 ISSN 0044-3468 C 21274F ZEITSCHRIFT FÜR SÄUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Irgan der Deutschen Gesellschaft für Säugetierkunde amı. g 18 T TE ISu0 U | { Johannesson-Gross, Kristina: Lernversuche in einer Zweifachwahlapparatur zum Hell-Dunkel-Sehen Hs Maul- wurfs (Talpa europaea L.). — Brightness discrimination of the mole (Taga ge: N in. 1 learning. experiments appliying a modified tube-maze method Fran 2 Re 193 Pen el ange Bastian, H. V.: Vorkommen und Zug der Rauhhauffledermaus (Pipistrellus nathusii Keyserling und Blasius, 1839) in Baden-Württemberg. — The occurrence and migration of Nathusius’ pipistrelle (Pipistrellus nathusii Keyserling & Blasius, 1839) in Baden-Württemberg 202 Bruorton, M. R.; Perrin, M. R.: The anatomy of the stomach and caecum of the Samango monkey, Cercopithecus mitis erythrarchus Peters, 1852. — Die Anatomie von Magen und Blinddarm der Diadem-Meerkatze Cercopithe- cus mitis erythrarchus Peters, 1852 210 Le Berre, M.; Le Guelte, L.: Structure de l’espace et retour au nid chez la gerbille de Mongolie (Meriones unguiculatus). — Space utilization and homing in the Mongolian gerbil (Meriones unguiculatus). -— Raumnutzung und Rückkehr zum Nest bei der Mongolischen Rennmaus (Meriones unguiculatus) 225 Bowland, A. E.; Perrin, M. R.: The effect of fire on the small mammal community in Hluhluwe Game Reserve. — Der Einfluß von Bränden auf die Kleinsäuger im Hluhluwe-Wildreservat 235 Patterson, |. J.: Responses of Apennine chamois to human disturbance. — Reaktionen apenninischer Gemsen auf menschliche Störung 245 Wissenschaftliche Kurzmitteilung Waerebeek, K. Van; Reyes, J. C.: First record of the Pygmy killer whale, Feresa attenuata Gray, 1875 from Peru, with a summary of distribution in the eastern Pacific.- Erster Nachweis eines Zwergschwertwales, Feresa attenuata Gray, 1875, von Peru und eine Zusammenfassung über die Verbreitung im östlichen Pazifik 253 Buchbesprechung 256 %Y Verlag Paul Parey Hamburg und Berlin HERAUSGEBER/EDTLORS P. J. H. van BREE, Amsterdam - W. FIEDLER, Wien -— H. Frick, München —- W. HERRE, Kiel - H.-G. Kıös, Berlin - H.-]J. Kunn, Göttingen - E. KuLzeEr, Tübingen - B. LAnza, Florenz — J. NIETHAMMER, Bonn — H. ReıcHstein, Kiel - M. RönHrs, Hannover - D. STArRcK, Frankfurt a. M. - F. Strauss, Bern -— E. THEnIUs, Wien — P. VOGEL, Lausanne SCHRIETLEITUNG/EDITORTATOFHIE@E H. SCHLIEMANN, Hamburg — D. Kruska, Kiel This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con- tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information Die Zeitschrift für Säugetierkunde veröffentlicht Originalarbeiten und wissenschaftliche Kurzmittei- lungen aus dem Gesamtgebiet der Säugetierkunde, Besprechungen der wichtigsten internationalen Literatur sowie die Bekanntmachungen der Deutschen Gesellschaft für Säugetierkunde. Verantwort- licher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Harald Schliemann. Zusätzlich erscheint einmal im Jahr ein Heft mit den Abstracts der Vorträge, die auf der jeweiligen Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde gehalten werden. Sie werden als Supplement dem betreffenden Jahrgang der Zeitschrift zugeordnet. Verantwortlich für ihren Inhalt sind ausschließlich die Autoren der Abstracts. Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z. Hd. Prof. Dr. Dieter Kruska, Institut für Haustierkunde, Biologie-Zentrum, Neue Universität, Olshausenstr. 40-60, D-2300 Kiel. Für die Publikation vorgesehene Manuskripte sollen gemäß den ‚„Redaktionellen Richtlinien‘ abgefaßt werden. In ihnen finden sich weitere Hinweise zur Annahme von Manu- skripten, Bedingungen für die Veröffentlichung und die Drucklegung, ferner Richtlinien für die Abfassung eines Abstracts und eine Korrekturzeichentabelle. Die Richtlinien sind auf Anfrage bei der Schriftleitung und dem Verlag erhältlich. Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der Korrekturfahnen erfolgen. Vorbehalt aller Rechte: Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei- ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grund- sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber- rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sınd mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of the first page of an article in this journal indicates the copyright owner’s consent that copies of the article may be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective, or for resale. For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the GEE: - Mit einer Beilage des Verlages Paul Parey Fortsetzung 3. Umschlagseite © 1988 Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstr. 12, D-2000 Hamburg 1; Lindenstr. 44-47, D-1000 Berlin 61. -— Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., D-2240 Heide/Holst. Z. Säugetierkunde 53 (1988) 193-201 \ © 1988 Verlag Paul Parey, Hamburg und Berlin N ISSN 0044-3468 % Nr len. ln, Lernversuche in einer Zweifachwahlapparatur enter zum Hell-Dunkel-Sehen des Maulwurfs (Talpa europaea L.) Von KRrIsTINA JOHANNESSON-GROSS Aus dem Fachbereich 19 (Biologie/Chemie) der Universität Kassel Eingang des Ms. 15. 05. 1987 Abstract Brightness discrimination of the mole (Talpa europaea L.) in learning experiments applying a modified tube-maze method Studied was the light-dark discrimination of the mole (Talpa europaea) ın learning experiments applying a special maze method. 8 animals (caught outside and tamed before the experiment) were trained to the light side of a modified tube-maze. The results were shown in quantity in learning curves. It became evident that moles can learn to distinguish between light and dark. 6 moles learned the task at a degree of illumination of 350 lux on the light (reinforced) side. 2 other animals reached the criterion level (90 % correct responses in 30 trials) not before an increase in ıllumination to 500 lux being the next phase of the experiment. The initially offered degree of illumination of 60 lux proved too low as a sign of discrimination. There were no temperature differences between the lıght and the dark side of the two-choice-training apparatus. And there was no indication whatever that moles find their way by means of infrared radiation. The possible role of the perception of light impulses from the mole in life conditions, especially the ecological function of the optical sense is discussed. Einleitung Talpa europaea lebt weitgehend unterirdisch in einem selbstgegrabenen und in großen Teilen dauerhaften Tunnelsystem (GODFREY und CROWCROFT 1960; QUILLIAM 1966b; MELLANBY 1974). Seine subterrane und fossoriale Lebensweise bedingt mannigfaltige anatomische und physiologische aber auch verhaltensmäßige Sonderanpassungen. So darf man bei ihm eine spezielle Lerndisposition zur Raumorientierung erwarten. Auf diesem Gebiet sind unsere Kenntnisse aber äußerst lückenhaft, systematische experimentelle Forschungen liegen kaum vor. Von den bisher untersuchten Sinnessystemen des Maulwurfs (vgl. QuiLLıam 1966c, d) hat sich der Tastsınn als besonders hoch entwickelt erwiesen. Ihm wird eine große Bedeutung für das Zurechtfinden in den verzweigten dunklen Gängen beigemessen (u.a. Krıszar 1940a, b, c; GopET 1951; QuitLLiam und ArMSTRONG 1963). Lernversuche zur taktilen Orientierung beim Wegelernen in Labyrinthen haben diese Auffassung experimen- tell bestätigt (CHRZANOWwSKI 1972; JOHANNESSON-GROSsS 1984, 1986). Im Gegensatz zum Tastsinn ist der Gesichtssinn der Talpıdae in unterschiedlichem Ausmaß rückgebildet. Unter den Altweltmaulwürfen (zu den Verhältnissen bei den Scalopinae siehe SLONAKER 1902; Lewis 1983) weist das Auge von Talpa europaea noch den geringsten Reduktionsgrad auf. Wie licht- und elektronenmikroskopische Untersu- chungen erkennen lassen, besitzt es alle für ein Sehen notwendigen Strukturen, zeigt aber deutliche Abweichungen vom Bau eines typischen Säugetierauges (Ciaccıo 1875; KoHL 1893, 1895; RocHoNn-DUVIGNEAU 1943; HENDERSON 1952; QUILLIAM 1966a; SIEMEN 1976). Nicht immer ist das winzige Auge des Maulwurfs (® ca. 1 mm) sichtbar. Meistens liegt es von Haut überdeckt unter dem Fell verborgen. In manchen Situationen tritt es jedoch aus der Lidspalte heraus und kommt zum Vorschein (Abb. 1). Dieses Phänomen ist verschiedenen Autoren zwar aufgefallen (SCHEFFER 1949; TusQues 1954; u.a.), trotzdem U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5304-0193 $ 02.50/0 194 Kristina Johannesson-Gross wurde Talpa europaea aber weithin für ein blin- des Tier gehalten (z.B. Su- zukı und Kurosomi 1972; PEvET et al. 1981) oder zu- mindest für ein solches mit nur wenig ausgeprägtem optischen Leistungsver- mögen (MELLANBY 1982). Erste verhaltensexperi- mentelle Ansätze zur Prü- fung des Gesichtssinnes dieser Tierart finden sich bei Krıszar (1940a) sowie bee Lunp und Lunp (1965, 1966). Während Krıszar eine Dressur des Maulwurfs auf ein Lichtsi- Abb. 1. Maulwurf mit hervortretendem Auge beim Verlassen einer gnal nach der Methode der Role klassischen Konditionie- rung nicht gelang, konn- ten Lunp und Lunn (1965) durch operantes Konditionieren ein positives Ergebnis erzielen. Allerdings war der lediglich von 4 Tieren erreichte Lernerfolg ausgesprochen instabil, und große individuelle Schwankungen der Lerndauer traten auf. Da außerdem in den entsprechenden Publikationen keine Lernkurven vorgestellt wurden, war dies Anlaß, weiterführende Lernexperimente mit Talpa enropaea zum Problemkreis des Hell-Dunkel- Sehens anzustellen und zu prüfen, ob und ın welchem Ausmaß sıch diese Tiere nach Helligkeitsdifferenzen räumlich orientieren können. Material und Methode Zur Untersuchung des optischen Leistungsvermögens wurde eine maulwurfsadäquate - im Hinblick auf die spezifische Fragestellung erweiterte - Röhrenlabyrinthmethode entwickelt, mit der es möglich war, durch Futterbelohnung positiv verstärkte Zweifachwahlen auf Hell (belohnt) gegen Dunkel (unbelohnt) durchzuführen. Zum Prinzip der Methode siehe Abb. 2 und JOHANNESSON-GROSsS und Gross (1982). Die Lernapparatur ist aus farblos-transparenten Kunststoffteilen zusammengesetzt. Sie besteht aus dem Entscheidungsraum (einem nach oben und unten offenen Plexiglaskasten) und aus insgesamt 3 PVC-Röhren. Eine dieser Röhren führt als Eingang in den Entscheidungskasten hinein. Ihr gegenüber liegt die Wahlseite der Apparatur mit 2 ebensolchen Röhren als Ausgängen (Abb. 3). Durch Überstülpen zweier Abdeckelemente kann eine der beiden Ausgangsröhren lichtundurchlässig gemacht werden. Die andere Ausgangsröhre bleibt hell und stellt so die Richtigwahlseite dar (Abb. 3b). Als zusätzliche visuelle Orientierungshilfe dienen 2 gegeneinander austauschbare Schieber. Sie sind zwischen Kastenaußenseite und Röhrenbeginn eingesetzt, jeweils fast die Hälfte der Seite einnehmend (Abb. 3a, b). Der eine Schieber ist mit weißer, der andere mit schwarzer d-c-fix-Folie beklebt. Darüber befindet sich zusätzlich ein Überzug aus gleichartiger durchsichtiger Schutzfolie. Zu Beginn eines jeden Durchganges befindet sich das Versuchstier (VT) im Startelement (Abb. 3b). Wird dieses geöffnet, läuft das VT bis zum Entscheidungsraum vor und gelangt über eine 3 cm hohe Stufe in diesen hinein. Um von dort den Lauf in der dunklen (falsch) oder hellen (richtig) Röhre fortzusetzen, muß das VT noch einmal eine 3 cm hohe Stufe überwinden. Diese Stufen (wie der ganze Entscheidungskasten überhaupt) sollen durch Anhalten bzw. Verzögern des Laufes die Wahrschein- lichkeit des Beachtens visueller Reize durch die VT erhöhen. Erreicht der Maulwurf das Zielelement, wird es hinter ıhm verschlossen, das Belohnungsloch geöffnet und eine Mehlkäferlarve als positive Verstärkung geboten. Ist diese gefressen, wechselt das VT in das Pausenelement über, von wo es vor Beginn des nächsten Durchganges erneut in das Startelement gelassen wird (Abb. 2). Pro VT fanden täglich 10 Durchgänge nacheinander statt. Zum Ausschließen geruchlicher Lernversuche zum Hell-Dunkel-Sehen des Maulwurfs 195 Abb. 2. Schema zum Versuchsablauf. Ab- holen eines Maulwurfs vom Haltungskäfig und Transport zum Startelement der Ver- suchsapparatur (1, 2); insgesamt 10 Durch- gänge (= Läufe) hintereinander werden von jedem Tier während einer Lernsitzung ab- solviert (3, 4); Rücktransport zur Käfigan- lage (5). Zwischen den Haltungskäfigen und der Versuchsanlage besteht durch die Transport- und Pausenelemente ein räumli- ches Kontinuum, keines der Tiere muß während der Versuche angefaßt werden Einflüsse wurden die Röhren nach jedem Durchgang mit reinem Alkohol ausge- wischt, und die Fließpapierunterlage (Abb. 3b) wurde gewechselt. Neonröhren und 3 Zusatzlichter in Form von 100-Watt-Glüh- lampen (Abb. 3b) leuchteten die Wahlap- paratur gleichmäßig aus. Durch Variieren des Abstandes von L 3 zur Anlage konnte die Beleuchtungsstärke der belohnten Seite und damit der Hell-Dunkel-Unterschied zwischen den beiden Ausgangsröhren ein- gestellt und schrittweise verändert werden. L1 und L2 blieben in ıhrer Position un- verändert. Die belohnte Seite der Zwei- ee | || fachwahlapparatur wurde für jeden Durch- Transport- Transport- gang nach Zufall bestimmt, wobei aber die- ne Senen! selbe Seite höchstens an 2 aufeinanderfol- genden der insgesamt 10 Durchgänge die belohnte sein konnte. Temperaturunter- L l schiede zwischen den beiden Wahlseiten waren nicht feststellbar (Sekunden-Ther- mometer mit Oberflächenmeßfühler Tech- notherm Typ 3000, Ablesegenauigkeit + 1/ 10°C). Als Lernparameter fand die prozen- tuale Häufigkeit der Richtigläufe (RL) in 10 Durchgängen Verwendung. Das Lernkrite- rıum galt als erfüllt, sobald vom VT durchschnittlich 90 % RL an 3 aufeinanderfolgenden Tagen erreicht waren. Getestet wurden 8 Wildfänge der Art Talpa enropaea verschiedenen Alters und Geschlechts, die zu Versuchsbeginn alle handzahm waren. Spontantendenzen der Seitenwahl waren nicht nachweisbar. Haltungskäfige Ergebnisse Entsprechend der schrittweisen Änderung der Beleuchtungsbedingungen auf der belohn- ten Wahlseite der Lernapparatur bestand die Dressur auf Hell-Dunkel-Unterschiede aus 3 Versuchsteilen. Versuchsteil I: Die Beleuchtungsstärke auf der belohnten Seite betrug 60 Lux. Unter diesen Bedingungen bleibt die durchschnittliche prozentuale Häufigkeit der RL aller VT (T 1 bis T 6) vom 1. bis zum 60. Tag ım Zufallsbereich (Abb. 4a). Bei keinem einzelnen der 6 Tiere war eine Bevorzugung der belohnten (hellen) Seite feststellbar. Versuchsteil I: Die VT wurden in der gleichen Weise wie in dem unmittelbar vorausgegangenen Versuchsteil I getestet, jedoch unter veränderten Beleuchtungsverhält- nissen. In dem 48tägigen Versuchsteil II war die Beleuchtungsstärke der belohnten Seite auf 350 Lux erhöht. 3 der bereits im Versuchsteil I untersuchten Tiere erreichten diesmal das Lernkriterium, und zwar nach 19 (T 2) bzw. 28 (T 1 und T 4) Tagen (Abb. 4b). Ein VT dieser Gruppe (T 3) starb vor Erreichen des Lernkriteriums. Wie die sequenzanalyti- sche Auswertung aber zeigt (Abb. 5), wird von diesem Maulwurf nach dem 93. Lauf im Kristina Johannesson-Gross 196 39UFF098 Sunuyoppgasnn-T spe sAreJI>FeNTyaW 9UL2 usIaIquy wmz Inu PIIM so ‘y9ofsdunuyopg = g TUIDPUS USPUSUSIYOY USSNIEMYONI USP UOA UN 76 "Mzq UI G/] sFewas ayIeIssgungysnajag uUSpuajjaisnzurs 19p ısı pun usyury uoA nyereddy ap 391yO9n9[2Q € 7 'sne Zipewy9ra]s UIOA UOA 9s9Ip uaIyanaJ pun Iydergadue arasjyey Pp Iyundpnm woA (WI 06) pueisqy woIuelsuoy Jruı purs 7 pun I 1 ‘(opragdıaasey)) SunaysrasppefursgystT us1ap pun uodweynfg-HeM-00T :E T ‘TI ‘LT (9) O3ejuesysnsza\ 93rpueisjoA aıp ne 1yDIspnY/ :519984 ‘(e) uıseysdunprsysssug usp ne 1ysıspny adeayos :syuı7 “ınzeseddejyemyseromZz 19p Sunuyarszeusyag 'g 'gqW 21, w>o6 yıpıs w206 „1.1 JU3WSJP1UDIS Juawasj3}Jodsunpı] aıdpdyaı) 31U3WaJ2YI3PpqY DM oO (8) 3 (gı9Mm) 49A91U3 N qalyas o [@) (zıpmua2s) 3 „uU9WOSJIJU9ISNDIJ SEE EZ ZUDESZ Lernversuche zum Hell-Dunkel-Sehen des Maulwurfs 197 100 = p= 0,05 n=6. (KT biste) & S Ho} rn) = ° x I: Oo 45 50 Tagle 60 100 Ar r j 2 PA ya a X/ RN IH V. O5 % b 50 (37 = Io) o e oO Ba 1 5 10 15 20 25 30 35 ade 48 61 70 80 90 ı0 79° 108 100 Richtiglaufe (%) & S AD) [) E © 2 1 5 10 15 20 25 30 35 40 45 50 55 60 T 68 61 70 80 90 100 110 120. Ko sie2178 Abb. 4. Kurven der Richtigläufe (RL) für die Dressur auf Hell-Dunkel-Unterschiede bei verschiede- nen Beleuchtungsstärken: Versuchsteil I (60 Lux), Versuchsteil II (350 Lux) und Versuchsteil III (500 Lux). (a) Mittelwertskurve der RL von 6 Maulwürfen (T 1 bis T 6) während des Versuchsteils I; angegeben ist die Standardabweichung. (b) Einzellernkurven von 3 Maulwürfen (T 1, T 2, T 4), die bereits Versuchsteil I absolviert haben. (c) Einzellernkurven von 2 bisher versuchsunerfahrenen Maulwürfen (T 7, T 8). (d) Einzellernkurven der Maulwürfe T 5 und T 6, deren Dressur wie bei T 1, T2 und T4 mit Versuchsteil I begonnen hat. Gepünktelte Linie: Signifikanzgrenze (p= 0,05: Prüfung einer Grundwahrscheinlichkeit nach KoLLEr 1969) 198 Kristina Johannesson-Gross Versuchsteil II die Annahmegrenze (90 % RL, p = 0,05) für das Eintreten des Lernerfol- ges erreicht. Das bedeutet, auch für den Maulwurf T 3 ist das Ergebnis der Dressur auf Hell-Dunkel-Unterschiede positiv zu werten. 2 bisher versuchsunerfahrene Tiere (T 7 und T 8) benötigten bis zum Erreichen des Lernkriteriums 25 (T 7) bzw. 39 (T 8) Tage (Abb. 4c). Versuchsteil III: Die Beleuchtungsstärke auf der belohnten Seite betrug im Anschluß an den Versuchsteil II vom 49. (Abb. 4c) bzw. 109. Tag (Abb. 4d) an 500 Lux. 2 Maulwürfe (IT 5 und T 6), die als einzige die Aufgabe bisher nicht gelernt hatten, erfüllten unter diesen Beleuchtungsverhältnissen rasch das Lernkriterium, der eine (T 6) nach 7, der andere (T 5) nach 10 Tagen (Abb. 4d). Die Lernkurven derjenigen VT, welche im Versuchsteil II das Lernkriterium bereits erreicht hatten, näherten sich jetzt in der Anzahl der RL dem 100 %-Niveau an (T 7 und T 8 in Abb. 4c). BI 70 eo on Diskussion 120 Läufe Trotz zahlreicher Untersuchungen, die zu einer weitgehenden Strukturaufklärung des Maulwurfsauges geführt haben, ist auf ex- perimenteller Grundlage bislang kaum et- was darüber bekannt, ob Talpa europaea zu Sehleistungen fähig ist und gegebenenfalls den optischen Sinn situationsspezifisch ein- setzt. In der vorliegenden Arbeit war es mög- lich, mit Hilfe einer speziellen Zweifach- wahlapparatur nach der Vorgehensweise ei- ner operanten Konditionierung Maulwürfe auf Hell-Dunkel-Unterschiede zu dressie- ren. Wie die Lernkurven zeigen, kann Tal- pa europaea Helligkeitsdifferenzen wahr- nehmen und diese — wenn sie nicht zu 60 Läufe Abb. 5. Lernkurve des Maulwurfs T 3 (sequenz- analytische Darstellung der Richtig- und Falsch- läufe nach HARrDER et al. 1972), der nach dem 12. Tag (120. Lauf bzw. Durchgang) im Ver- suchsteil II gestorben ist. Unten: Lauf 1 bis 60; oben: Lauf 61 bis 120 gering sind - in der spezifischen Lernsitua- tion als Orientierungsmerkmale nutzen. Im Gegensatz zu Lunp und Lunp (1965) ist hier in den Lernkurven eine stabile Kann- phase ausgebildet (Abb. 4b bis c). Die zur optischen Orientierung erforderliche Unterschiedsschwelle ist individuell verschieden; 6 Maulwürfe erlernten die Aufgabe beı einer Beleuchtungsstärke von 350 Lux auf der hellen Seite, 2 Tieren war dies erst nach Erhöhung der Beleuchtungsstärke auf 500 Lux möglich (Abb. 4d). Eine Beleuchtungsstärke von 60 Lux erwies sıch als unterschwellig und konnte von keinem der getesteten Maulwürfe als Orientierungszeichen eingesetzt werden (Abb. 4a). Hinweise auf eine Orientierung nach Infrarotstrahlen, wie sie für die Vampirfleder- maus Desmodus rotundus nachgewiesen ist und in ihrer Bedeutung diskutiert wurde (SCHMIDT und ManskE 1982; KÜRTEN und ScHMmipr 1982), ergaben sich für Talpa europaea ın diesen Versuchsreihen nicht. Offen bleibt die Frage nach der Funktion des optischen Sinnes von Talpa europaea. Es ist bekannt, dafß der europäische Maulwurf einen Teil seiner Aktivitätsphasen oberirdisch verbringt (MoRrrıs 1966). Zahlreiche zufällige Einzelbeoachtungen verdeutlichen ın ihrer Gesamtheit, daß diese Tiere regelmäßig zu jeder Jahres- und Tageszeit ihren unterirdischen Bau verlassen und auf der Erdoberfläche tätig werden (u.a. Remus 1901; HAUCHECORNE 1927; SCHAERFFENBERG 1940; HoRNUNG 1942; NIETHAMMER 1963). Sie weichen vor Überschwemmungen ihrer Tunnelsysteme (teils schwimmend) auf höher gelegene Gelän- Lernversuche zum Hell-Dunkel-Sehen des Maulwurfs 199 deabschnitte aus und kehren nach Rückgang der Fluten rasch wieder zu ihren ehemaligen Wohngebieten zurück (CstzmAzıa 1982; JOHANNESSON-GROSss und Gross 1986). Auch ihr Nachweis im Beutespektrum von Greifen und Eulen bestätigt indirekt ebenfalls Oberflächenaktivitäten (z.B. SOUTHERN 1954; SKOCZEN 1962). Welche Sinne den Maulwurf außerhalb seines Baues leiten, ist weitgehend unbekannt. Gehör, Geruch, Erschütterungs- und Tastsinn werden als vorrangig betrachtet (HERTER 1957). Eine optische Orientierung scheint -— wenn überhaupt - von untergeordneter Bedeutung zu sein. Dies schon deswegen, da aus anatomischen Gründen ein Verlust des Bildsehens angenommen werden muß. Diejenigen Teile des Auges, die zum Hervorbrin- gen eines scharfen Bildes auf der Retina notwendig sind (Cornea, Linse, Glaskörper), werden von ihrer Ausbildung her dieser Aufgabe nicht mehr gerecht (SoKoLOwA 1964; QUILLIAM 1966c). SIEMEN (1976) weist in diesem Zusammenhang zusätzlich auf Degenera- tionserscheinungen der Netzhaut hin. Der optische Sinn vieler Säugetiere ist aber nicht nur für ein Zurechtfinden im Raum wichtig, ebenso bedeutungsvoll ist er für die Orientierung in der Zeit (HOFFMANN 1981). Maulwürfe lassen einen 24-Std.-Rhythmus ihrer lokomotorischen Aktivität erkennen, der zusätzlich von Kurzzeitzyklen überlagert sein kann (z.B. GODFREY 1955; MELLANBY 1967; MEESE und CHEESEMAN 1969). Ergebnisse einer über 12 Monate durchgeführten Studie zur Tagesaktivität von Talpa europaea durch Woops und MEAD-BRIGGS (1978) legen die Vermutung nahe, daß die auftretenden Aktivitätsschübe vom täglichen Lichtgang und seiner Änderung im Jahresverlauf beeinflußt werden. Aber nicht nur hinsichtlich tagesrhythmischer Vorgänge scheint das Licht bei diesen Tieren eine Rolle zu spielen, sondern möglicherweise auch bei der zeitlichen Einnischung des Fortpflanzungsverhaltens in das Jahresgeschehen. Talpa europaea ist eine monöstrische Art mit einer nur relativ kurzen Brunftzeit im Frühjahr (MATTHEws 1935; DEANESLY 1966). Auch wenn man unterstellt, daß dem Fortpflanzungszyklus des Maulwurfs ein circannualer Rhythmus zugrunde liegt, erscheint die Photoperiode in den gemäßigten Breiten mit ihrer störungs- freien Information als Zeitgeber zur Synchronisation der Brunftzeit geeignet und notwen- dig. Der optische Sinn von Talpa europaea mag also weniger der räumlichen als vielmehr der zeitlichen Orientierung dienen. Die hierzu erforderliche Kontrolle des Lichtganges ım Tages- und Jahresverlauf ıst den Tieren möglich; sie sind, wie ın den Lernversuchen gezeigt werden konnte, zur Lichtperzeption fähig, und sıe können diese Fähigkeit auch nutzen, da sie sowohl in ihren Gängen oberflächennah vorkommen als diese auch regelmäßig verlas- sen, um oberirdisch aktiv zu werden. Zusammenfassung Untersucht wurde die Fähigkeit zur Hell-Dunkel-Diskrimination von Talpa europaea L. Mit einer speziellen Lernapparatur in Form eines modifizierten Röhrenlabyrinthes war es möglich, durch Futterbelohnung positiv verstärkte Zweifachwahlen auf Hell (belohnt) gegen Dunkel (unbelohnt) durchzuführen. Wie die Lernkurven zeigen, ıst Talpa europaea fähig, Heiligkeitsunterschiede wahr- zunehmen. Sind sie nicht zu gering, werden sie ın der Lernsituation als Orientierungsmerkmal eingesetzt. Von 8 VT (Wildfänge verschiedenen Alters und beiderlei Geschlechts, die zu Versuchsbe- ginn handzahm waren) erlernten 6 Maulwürfe die Aufgabe bei einer Beleuchtungsstärke von 350 Lux auf der hellen (belohnten) Seite, während 2 weitere Tiere das Lernkriterium (90% RL an 3 aufeinanderfolgenden Tagen) erst nach Erhöhung der Beleuchtungsstärke auf der belohnten Seite von 350 Lux auf 500 Lux erreichten. Welche Bedeutung der Lichtwahrnehmung bei der weitgehend unterirdischen Lebensweise von Talpa enropaea zukommt, wird diskutiert. Literatur CHRZANOWSKI, Z. (1972): Untersuchungen über die Orientierung von Maus und Maulwurf im Labyrinth. Warschau: Hochschule für Landwirtschaft. (In polnisch) Cıiaccıo, G. V. (1875): Osservazıoni intorno alla membrana del descemet e al suo endotelio con una descrizione anatomica dell’ occhio della Talpa europaea. Mem. Accad. Scı. Ist. Bologna, Ser. 3, 5, 501-516. 200 Kristina Johannesson-Gross CsızMAZIA, G. (1982): Die Wirkung des Hochwassers der Theiss auf die Population der Maulwürfe (Talpa europaea Linn€ 1758) im Inundationsraum bei Körtvelys (Ungarn). Tiscia 17, 219-224. Deranssty, R. (1966): Observations on reproduction in the mole (Talpa europaea). Symp. Zool. Soc. Lond. 15, 387402. GoDET, R. (1951): Contribution a l’ethologie de la taupe (Talpa europaea L.). Bull. Soc. Zool. Fr. 76, 107-128. GODFREY, G. K. (1955): A field study of the activity of the mole (Talpa europaea). Ecology 36, 678-685. GODFREY, G. K.; CROWCROFT, P. (1960): The life of the mole (Talpa europaea Linnaeus). London: Museum Press. | HARDER, W.; SCHÄFER, D.; SCHMIDT, U. (1972): Auswertung von Dressurversuchen (Zweifachwah- len) mit der Sequenzanalyse. Zool. Jb. Physiol. 76, 585-592. HAUCHECORNE, F. (1927): Studien über die wirtschaftliche Bedeutung des Maulwurfs (Talpa euro- paea). Z. Morphol. Okol. Tiere 9, 439-571. HENDERSoN, T. (1952): The eye of the mole. Brit. J. Ophthal. 36, 637. HERTER, K. (1957): Das Verhalten der Insectivoren. Handbuch der Zoologie, 8 (10): 1-50. Horrmann, K. (1981): Photoperiodism in vertebrates. In: Handbook of behavioral neurobiology. Bd. 4: Biological rhythms. Hrsg. J. Aschorr. New York, London: Plenum Press. pp. 449-473. Hornung, V. (1942): Der Maulwurf als Tagtier. Zool. Gart. 14, 104. JOHANNEsson-GRoss, K. (1984): Verhaltensbiologische Untersuchungen zum Thema Lernen am Maulwurf (Talpa europaea L., Insectivora, Talpıdae) mit Ausblick auf die zoodidaktische Bedeu- tung dieses Tieres. Diss. Gesamthochschule Kassel. JOHANNESSON-GROSsSs, K. (1986): Zur taktilen Orientierung des Maulwurfs (Talpa europaea L.): Lernversuche in einem Y-förmigen Röhrenlabyrinth. Verh. Dtsch. Zool. Ges. 79, 216-217. JOHANNEssoN-GRoSss, K.; Gross, H. (1982): Lernversuche mit Maulwürfen (Talpa europaea L.) unter Anwendung einer speziellen Labyrinthmethode. Z. Säugetierkunde 47, 277-282. JOHANNESSON-GRoSss, K.; Gross, H. (1986): Verlauf der Wiederbesiedelung einer Flußauenregion durch Maulwürfe (Talpa europaea L.) nach zeitweiser Überschwemmung. Z. angew. Zool. 73, 135-144. Kor, C. (1893): Rudimentäre Wirbelthieraugen. Zweiter Theil: Das Auge von Talpa europaea. Bibliotheca Zool. H. 14, 1-178. Kon, C. (1895): Rudimentäre Wirbelthieraugen. Dritter Theil: Zusammenfassung. Bibliotheca Zool. H. 14, 179-274. KOLLER, $. (1969): Neue graphische Tafeln zur Beurteilung statistischer Zahlen. Darmstadt: Stein- kopff Krıszat, G. (1940a): Untersuchungen zur Sinnesphysiologie, Biologie und Umwelt des Maulwurfs (Talpa europaea L.). Z. Morphol. Okol. Tiere 36, 446-511. . Krıszat, G. (1940b): Die Orientierung im Raume bei Talpa europaea. Z. Morphol. Okol. Tiere 36, 512-556. Krıszart, G. (1940c): Wie orientiert sich der Maulwurf in seinen Gängen? Umschau 44, 561-565. KÜRTENn, L.; SCHMIDT, U. (1982): Die Nasengruben der Vampirfledermaus Desmodus rotundus: Sinnesorgane zur Wahrnehmung von Wärmestrahlung. Z. Säugetierkunde 47, 193-197. Lewis, T. H. (1983): The anatomy and histology of the rudimentary eye of Neurotrichus. Northwest Scı. 57, 8-15. Lunp, R. D.; Lunp, ]J. $S. (1965): The visual system of the mole, Talpa europaea. Exptl. Neurol. 13, 302-316. Lunp, R. D.; Lunp, ]J. S. (1966): The central visual pathways and their functional significance in the mole (Talpa enropaea). J. Zool. 149, 95-101. MATTHEws, L. H. (1935): The oestrus cycle and intersexuality in the female mole (Talpa europaea Linn.). Proc. Zool. Soc. Lond. 1935, 347-383. MEESE, G. B.; CHEESEMANn, C. L. (1969): Radio-active tracking of the mole (Talpa europaea) over a 24-hour period. J. Zool. 158, 197-224. MELLANBY, K. (1967): Food and activity in the mole Talpa europaea. Nature 215, 1128-1130. MELLANBY, K. (1974): The mole. London: Collins. MELLANBY, K. (1982): Mysteries of the molehills. Country Life 171, 1144, 1147. MoRrıs, P. (1966): The mole as a surface dweller. J. Zool. 149, 46-49. NIETHAMMER, G. (1963): Notizen über den Maulwurf (Talpa europaea). Säugetierkdl. Mitt. 11, 79-80. PEvET, P.; Baremans, M. G. M.; REUVER, G. F. DE (1981): The pineal gland of the mole (Talpa europaea L.). VII. Activity of hydroxyindole-O-methyltransferase (HIOMT) in the formation of 5-methoxytryptophan, 5-methoxytryptamine, 5-methoxyindole-3-acetic acıd, 5-methoxytrypto- phol and melatonine in the eyes and the pineal gland. J. Neural Transmission 51, 271-282. QuirLıam, T. A. (1966a): The problem of vision in the ecology of Talpa europaea. Exp. Eye Res. 5, 63-78. QuiLLiam, T. A. (Ed.) (1966b): The mole: its adaption to an underground environment. J. Zool. 149, 31-114. Lernversuche zum Hell-Dunkel-Sehen des Maulwurfs 201 QuiLLıam, T. A. (1966c): The mole’s sensory apparaturs. J. Zool. 149, 76-88. QuiLLıam, T. A. (1966d): Unit design and array patterns in receptor organs. In: Touch, heat and pain. Ed. by A. V. $S. DEREucK; J. KnıGHT. Ciba Foundation Symposium. London: Churchill. 86116: en, T. A.; ARMSTRONG, J. (1963): Mechanorezeptoren. Endeavour 22, 55-60. Remus, C. (1901): Einige Beobachtungen aus dem Winterleben des Maulwurfs. Prometheus 12, 652-654. ROCHON-DUVIGNEAU, A. (1943): Les jeux et la vision des vertebres. Parıs: Masson. SCHAERFFENBERG, B. (1940): Die Nahrung des Maulwurfs (Talpa enropaea L.). Z. Säugetierkunde 14, 272-277. SCHEFFER, T. H. (1949): Ecological comparisons of three genera of moles. Trans. Kansas Acad. Sci. 52, 30-37. SCHMIDT, U.; MAnsKE, U. (1982): Thermopräferenz bei der Gemeinen Vampirfledermaus (Desmodus rotundus). Z. Säugetierkunde 47, 118-120. SIEMEN, D. (1976): Elektronenmikroskopische Untersuchungen zur Reduktion des Auges bei unterir- disch lebenden oder nachtaktiven Säugetieren (Talpa europaea Linng, 1758, Erinaceus enropaeus Linne, 1758, Echinops telfaırı Martin, 1838). Dipl.-Arbeit, Univ. Kiel. SKOCZEN, S. (1962): Age structure of skulls of the mole Talpa europaea Linnaeus 1758 from food of the buzzard (Buteo buteo L.). Acta Theriol. 6, 1-9. SLONAKER, J. R. (1902): The eye of the common mole Scalopus aquaticus machrinus. J. comp. Neurol. 12, 335-366. SOKOLOWA, Z. A. (1964): Reduction of the eye in moles as a result of function limitation. Folıa Biol. 12, 183-201 (in russ. mit engl. Zus.). SOUTHERN, H. N. (1954): Tawny owls and their prey. Ibis 96, 384-410. Suzukı, H.; Kurosom1, K. (1972): Fine structure of the cutaneous nerve endings in the mole snout. Arch. histol. Jap. 34, 35-50. Tusaugs, J. (1954): Ouverture palpebrale et developpement du globe oculaire sous l’action de la thyroxine chez la taupe (Talpa europaea L.). C. R. Acad. Scı. 238, 2562-2564. Woops, J. A.; MEAaD-Brıccs, A. R. (1978): The daily cycle of activity in the mole (Talpa europaea) and its seasonal changes, as revealed by radioactive monitoring of the nest. J. Zool. 184, 563-572. Anschrift der Verfasserin: Dr. KrısTINA JOHANNESSON-GROSsS, Kimpelstraße 8, D-3500 Kassel, FRG Z. Säugetierkunde 53 (1988) 202-209 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Vorkommen und Zug der Rauhhautfledermaus (Pipistrellus nathusii Keyserling und Blasius, 1839) in Baden-Württemberg Von H. V. BasTIan Institut für Zoologie III, Universität Tübingen, Abteilung für Physiologische Ökologie Eingang des Ms. 15. 5. 1987 Abstract The occurrence and migration of Nathusius’ pipistrelle (Pipistrellus nathusü Keyserling & Blasins, 1839) in Baden-Württemberg During the whole year Nathusius’ pipistrelle can be found in southwest Germany, but not one record of reproduction is known from Baden-Württemberg. Pipistrellus nathusü is much more frequent in autumn and late summer (about 85 % of all records) than in other seasons. Therefore it is supposed, that this species is normally migratory in southwest Germany and hibernates there only seldom, but more in southern countries (e.g. France). It could be shown that Nathusius’ pipistrelle arrıves 10 days earlier at the north eastern part of Baden-Württemberg than at Lake Konstanz and about four weeks earlier than at the Regio Basiliensis. A possible migration route is discussed. Einleitung Die Analyse saisonaler Wanderungen von Fledermäusen ist durch die nächtliche und oft versteckte Lebensweise und durch die Seltenheit der Tiere erschwert. Rauhhautfledermäuse überwintern in der Regel nicht in Felshöhlen und sind dadurch gezwungen, weiträumige Wanderungen durchzuführen (RoER 1971), die sie in Gebiete leiten, deren durchschnittliche Tagestemperaturen im Januar -6°C nicht unterschreiten (STRELKOV 1969). Der Verbreitungsschwerpunkt der Art liegt im zentralen und südlichen Rußland (STRELKOV 1969; ROER 1974; HANAK und GAISLER 1976; CLAUDE 1976), von wo Wanderungen zum Kaukasus und auf den Balkan bis nach Griechenland belegt sind (STRELKOV 1969). Untersuchungen über das Vorkommen in den westlichen Randgebieten liegen aus der DDR (Heise 1982), der Tschechoslowakei (HAnaK und GAISLER 1976) und aus Österreich (BAUER und WIRTH 1979) vor. Für andere mitteleuropäische Länder stellten Roer (1974, 1976) und Kock und SCHwARTING (1987) zahlreiche Nachweise zusammen. In der vorliegenden Arbeit wird versucht, das Vorkommen der Rauhhautfledermaus in Südwest-Deutschland zu beschreiben und aus der Verteilung der Funddaten Hinweise auf Wanderungen dieser Art zu erhalten. Unterschiedliche Zugzeiten von Männchen und Weibchen, wie sie von STRELKOV (1969) und HEıseE (1982) vermutet werden, wurden nicht berücksichtigt. Material und Methoden Für die Auswertung wurden die Publikationen von CLAUDE (1976), StuTz und HAFFNER (1985) sowie die Daten der Fledermauskartierung in Baden-Württemberg (KuLzer et al. 1987) herangezogen. Zum Vergleich dazu wurden Mitteilungen aus anderen Gebieten Deutschlands berücksichtigt (Tab. 1). Es standen Daten von 135 Fundorten mit insgesamt 508 Tieren zur Verfügung. 326 Tiere wurden alleın aus Baden-Württemberg und dem Deutsch-Schweizer Grenzgebiet gemeldet. Zur Darstellung der Wanderung wurde Baden-Württemberg in vier Sektoren eingeteilt, deren Grenzen sich an geographischen Merkmalen (Schwarzwald, Schwäbische Alb) und an Orten mit Fundhäufungen orientierten (Abb. 1). Es wurden nur Funde gewertet, von denen mindestens der U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5304-0202 $ 02.50/0 Vorkommen und Zug der Rauhhautfledermaus in Baden-Württemberg 203 Tabelle 1. Anzahl der Funde von Rauhhautfledermäusen in verschiedenen Gebieten Deutschlands Gebiet Anzahl Fundorte Anzahl Tiere Quelle Baden-Württemberg und 91 326 CLAUDE 1976; GEBHARD 1983; Deutsch-Schweizer Grenz- MÜLLER und WIDMER 1983; gebiet StuTz und HAFFNER 1985; KuLzer et al. 1987 Rhein-Main-Gebiet Kock 1981; Kock und SCHWARTING 1987 Nordrhein-Westfalen RoeEr 1976; VIERHAUS und BüLow 1978 Schleswig-Holstein DIETERICH 1982 Summe Monat bekannt war. Der Erfas- k sungszeitraum erstreckte sich Yo für Baden-Württemberg von oN 1962 bis 1986. Die übrigen Fun- de gehen bis in das Jahr 1908 zurück. Für die Berechnung der Zugdaten wurde der Juli als er- Nee ster Monat gewertet (der Monat uiie mit den wenigsten Funden). ® B Für jeden Sektor wurde die Ge- HBe samtsumme der ın einem Monat gemeldeten Pipistrellus nathusü (ohne Berücksichtigung der 0Se He RB. Jahreszahl) sowie der Median 1 und das 95 %-Konfidenzinter- We vall aller Funde ermittelt. Für do die Berechnung der Median- Ee s SE werte wurden nur die Funde De 29 a 2 3 berücksichtigt, von denen das exakte Datum bekannt war. Die Fundverteilung im Sektor Ill war zweigipfelig, in allen ande- ren Fällen eingipfelig, so daß zur Berechnung des Medians ım ersten Fall nur die Funde in den Monaten Juli bis Dezember, sonst die des ganzen Jahres be- rücksichtigt wurden. Die in der vorliegenden Untersuchung be- rechneten Mediane sınd somit stets als Indices für Durchzugs- zeiten ım Herbst zu werten. Ergebnisse Jahreszeitliche Verteilung der Funde in Baden-Würt- —JI temberg In Baden-Württemberg und en De Schwerer Abb. 1. Aufteilung der Fläche Baden-Württembergs in vier Sekto- b; e ... ren (I-IV). Lage der zur Interpretation des Herbstzuges wichtigen Grenzgebiet UULde Pipi- Gebirge: 1= Harz; 2= Thüringer Wald; 3 = Erzgebirge; 4 = strellus nathusil ın allen Mo- Fränkische Alb; 5 = Schwäbische Alb; 6 = Schwarzwald 204 H. V. Bastian naten nachgewiesen. 84,6 % (= 215 Tiere) aller Nachweise entfielen jedoch auf die Monate September bis Februar. Nur 10 Tiere (= 3,9 %) wurden in der Zeit von Juni bis August gefunden (Tab. 2). Die Verteilung der Funddaten ist ungleichmäßig (x’= 209,8; p < 0,001). Der Median des vom 1.7 bis 30.6 gewerteten Jahres errechnet sich für Baden- Württemberg auf den 5. November (95 %-Konfidenzintervall: 25. 10.-21. 11.; Tab. 3). Das zweite schwach ausgebildete Fundmaximum im Spätwinter (Januar/Februar; Abb. 2) beeinflußt den Medianwert nicht. Die Befunde deuten auf eine saisonale Wanderung der Rauhhauttledermaus in Baden-Württemberg hin. Tabelle 2. Verteilung der in Baden-Württemberg gefundenen Rauhhautfledermäuse auf die vier Sektoren der Abb. 2 Bereich 12 Summe I II III IV Summe 3 D, 5 I = Nordbaden; II = Unterfranken und Neckar-Raum um Tübingen und Stuttgart; III = Südbaden und Raum Basel; IV = Bodenseeraum und Schweizer Grenzgebiet. Unberücksichtigt ist die große Männchengruppe aus Büsingen (Kreis Waldshut; MÜLLER und WIDMER 1983). SZ zZ N = 254 Baden -Würftemberg Abb. 2. Jahreszeitliche Verteilung der Rauhhaut- fledermausfunde in Baden-Württemberg. Aufge- tragen ist die Zeit von Juli bis Juni. Nicht be- rücksichtigt wurde eine große Männchengesell- schaft in einem Zwischenquartier ın Büsingen (MÜLLER und WIDMER 1983). |: Median; +41: 95 %-Konfidenzintervall oO INDIVIDUEN w un 3 oO _ oO Regionale Zugmuster Sowohl im Neckartal von Stuttgart und Tübingen, als auch im Bodenseeraum wurden in den Monaten September und Oktober die meisten Tiere gemeldet (Abb. 3). Das Neckartal erreichen die Rauhhautfledermäuse jedoch im Mittel 10 Tage früher als den Bodensee (p < 0,05; Median-Test; Tab. 3). Dort können die Tiere auch ganzjährig beobachtet werden, während im Neckartal außerhalb der Herbstmonate nur noch im Januar und Februar Tiere gemeldet wurden. Am Bodensee veranlaßt möglicherweise das warme Klima die Tiere zur Rast. Funde von bis zu 14 Tieren in den Monaten Dezember, Januar und Februar (Tab. 2) lassen zudem vermuten, daß Rauhhautfledermäuse dort auch überwin- tern. In der Region Basel ist die Anzahl der Funde in den Monaten November und März maximal. Der Zugmedian fällt auf den 7. 11. (95 %-Konfidenzintervall: 21. 10.-16. 11.; Tab. 3) und damit drei Wochen später als im Bodenseegebiet (p < 0,05; Median-Test; Tab. 3). Aus Nordbaden sind bisher noch keine Herbstnachweise bekannt. Die geringe Anzahl der Funde läßt keine Aussagen über jahreszeitliche Zugbewegungen der Art ın diesem Landesteil zu. Vorkommen und Zug der Rauhhautfledermaus in Baden-Württemberg 205 Abb. 3. Die Anzahl der je Monat mitgeteilten Rauhhautfledermäuse in den vier Sektoren Baden- Württembergs. In jedem Diagramm ist die Anzahl der berücksichtigten Funde (N), der Median des Herbstmaximums (|) und das 95 %-Konfidenzintervall (1) eingezeichnet Das Vorkommen in anderen Gebieten Deutschlands Umfangreiches Datenmaterial ist aus dem Rhein-Main-Gebiet bekannt (Kock 1981; Kock und SCHWARTING 1987), wo seit 1908 insgesamt 157 Pipistrellus nathusii beobachtet wurden. In der Regel fand man sie ın Nistkä- sten. Auch dort wurden im September die meisten Tiere festgestellt (70 % aller Funde, Abb. 4). Die restlichen Mitteilungen konzen- trierten sich nahezu zur Hälfte auf die Monate Januar und Februar. Nur 25 Tiere (= 15,9 %) fand man in der übrigen Zeit. Auch ein weite- rer Befund spricht für eine nur kurzzeitige Belegung der Quartiere. In den Kästen, in denen im September Rauhhautfledermäuse vorgefunden wurden, befand sich 2-3 Wochen vorher nicht einmal Fledermauskot (SCHWAR- TING, mündl. Mitt.). Abb. 4. Jahreszeitliche Verteilung von im Rhein- Main-Gebiet gefundenen Rauhhautfledermäusen (nach: Kock und ScHwARTInG 1987). Aufgetra- gen ist die Zeit von Julı bis Juni Tabelle 3. Medianwerte und 95 %-Konfi- denzintervalle der in den vier Bereichen Baden-Württembergs gefundenen Rauh- hautfledermäusen, bezogen auf das am 1. Juli beginnende Jahr Bereich Median 95 %-Konfidenz- x intervall So le 28.12.11. 310. 29. 9.-19. Fo ule 21. 10.-16. 18. 10. PF1N0=IG: Sealele 25. 10.-21. N N rl | NEN io =) Rhein-Main-Gebiet e& > 40 40 zu 20 JASOND!IJFMAM J MONAT 206 H. V. Bastian Nur wenige Daten liegen aus anderen Bundesländern vor. VIERHAus und BüLow (1978) berichten außer von 3 älteren Funden (je ein Tier im August, September und November) von 16 Rauhhautfledermäusen ın Nordrhein-Westfalen. Diese wurden an zwei September- tagen in Nistkästen angetroffen. Wenige Tage später waren alle Kästen leer. RoER (1976) beobachtete im November eine Rauhhautfledermaus an den Krickenbecker Seen (Kreis Viersen). Aus Schleswig-Holstein sind bisher 5 Funde in den Monaten September und Oktober bekannt (DIETERICH 1982). Alle Funde aus diesen Bundesländern stimmen mit den Fundhäufungen in Baden- Württemberg und dem Deutsch-Schweizer Grenzgebiet gut überein. Sie deuten ebenfalls auf einen Herbstzug der Rauhhautfledermaus in der Bundesrepublik hin. Diskussion Rauhhautfledermäuse sind zu weiträumigen Wanderungen fähig (STRELKOV 1969; ROER 1974, 1976; CLauDE 1976; DIETERICH 1982; Kock und SCHWARTINnG 1987), was bisher durch etwa 25 Fernfunde (>100 km) belegt ist. In Deutschland kann die Art in allen Monaten beobachtet werden, wenngleich sie im Herbst und Winter sehr viel häufiger ist (Abb. 2 und 4). Überwinternde Tiere sind wiederholt festgestellt worden (z.B. KuLzeEr et al. 1987), Wochenstuben sind dagegen bis auf eine Ausnahme (Isser et al. 1977) aus diesem Jahrhundert nicht bekannt. Auch andere Sommernachweise sind spärlich. Dagegen ist die Art ın Berlin und der DDR stellenweise sogar häufig (Schmipr 1977, 1978; Heıse 1982; Hızssch 1983). Die von KLAwITTER (1974) festgestellte hohe Dichte von Pipistrellus nathusii ın West-Berlin könnte jedoch gleichfalls ziehende Tiere betreffen, da Rauhhautfle- dermäuse auch hier besonders häufig im September gefunden wurden. In der DDR beringte Rauhhautfledermäuse zeigen eine deutliche SW bis SSW orien- tierte Wegzugrichtung, wıe durch Ringfunde aus der Bundesrepublik Deutschland, der Schweiz und Frankreich belegt ıst (Zusammenfassung bei Kock und SCHWARTING 1987). Die Rauhhautfledermäuse verlassen ihre Fortpflanzungsquartiere in den Monaten August bis Oktober; im Winter fehlen sie in der DDR weitgehend (SCHOBER 1960; HaenseL 1967; SCHMIDT 1978; Heise 1982). Nimmt man einen Wegzug ım September an, wobei die ? möglicherweise vor den d abziehen (STRELKOV 1969; HEısE 1982), könnten diese Fledermäuse in den Monaten September/Oktober ım Neckartal und am Bodensee eintreffen. Das weitgehende Fehlen in anderen Teilen der Bundesrepublik ließe sich mit einer angestammten Zugrichtung SW bis SSW und der Leitwirkung der Mittelgebirgszüge erklären. Die Funde ın den Niederlanden (WIJNGAARDEN et al. 1971) deuten jedoch auch auf einen, möglicherweise schwächeren Zug durch die norddeutsche Tiefebene hin. Bei der nach SW gerichteten Wanderung würden die meisten aus der DDR und Polen kommenden Rauhhautfledermäuse nach Umgehung des Harzes, des Thüringer Waldes und des Erzgebirges zwangsläufig in den Main-Neckar-Raum fliegen. Im SW schließt sich dann ein nach NNO geöffneter Gebirgstrichter an (im W der Schwarzwald, im O die Schwäbische und Fränkische Alb). Auf diesem hypothetischen Zugweg (Abb. 5) könnten die Rauhhautfledermäuse schließlich das klimatisch günstige Bodensee-Hochrhein-Gebiet erreichen. Im weiteren Zugverlauf müßten sich die Tiere dann SW oder W orientieren und dabei entweder durch das Schweizer Mittelland, entlang des Genfer Sees, des Rhöne-Tals nach Südfrankreich oder im weiteren Verlauf des Hochrheines durch die Burgundische Pforte in das Rhöne-Tal fliegen. Das im November beobachtete Zugmaximum in der Baseler Gegend macht den zweiten Zugweg wahrscheinlich. Widersprüchlich dazu sınd allerdings die Angaben von ArrrEn (1961), der die Tiere am Bergpaß Col de Bretolet hauptsächlich im September gefangen hat. Möglicherweise werden dort Tiere gefangen, die aus südlicheren Wochenstubenpopulationen stammen. Zu berücksichtigen ist bei allen Überlegungen, daß die Art entlang des gesamten Weges Vorkommen und Zug der Rauhhautfledermaus in Baden-Württemberg 207 überwintern kann, sowohl k in den Höhlen der Schwä- A bischen Alb (KuLzer et al. : 1987), als auch ın Baum- höhlen oder Nistkästen =iT2 (FELTEN und KLEMMER HH 1960; RoER 1976; Kock | ° 1981; Kock und SCHWAR- HBo B TING 1987). Dersariekzuer ın. dıe Sommerquartiere ist in der Funddatenverteilung nicht klar zu erkennen, da der Nebengipfel (Abb. 2) zwei Interpretationen zuläßt: Es kann sıch einerseits um Meldungen von Tieren handeln, die auf dem Heimzug in die Sommer- quartiere sind. Es könnte sich aber auch um eine durch äußere Bedingungen (z.B. Kälteeinbruch) ver- ursachte Häufung von Funden am Ort überwin- ternder Tiere handeln. In Sommerquartieren bei Halle wurden die er- sten Tiere Anfang April festgestellt (SCHOBER 1960); andere Autoren ge- ben als frühestes An- 100 km kunftsdatum in der DDR Anfang Mai an (SCHMIDT Abb. 5. Hypothetischer Zugweg der Rauhhautfledermaus aus po- 1978; Heıse 1982). Damit tentiellen Wochenstubenquartieren in der DDR und Polen. Ange- ER SE . nommen wird eine angeborene Herbstzugrichtung nach SW bis wird un „Frühjahrszug SSW, was von Ringfunddaten in dieser Richtung belegt wird. in Baden-Württemberg im Abkürzungen wie in Abb. 1 Januar unwahrscheinlich. Die relativ häufigen Funde könnten dagegen durch verschärfte Kälteperioden verursacht worden sein. Sie können beı überwinternden Tieren zur Unterbrechung des Winterschlafes und zur Suche nach geschützteren Quartieren führen. Gerade in Baumhöhlen überwin- ternde Fledermäuse sind dadurch sehr gefährdet (Kurzer 1986). Die Tatsache, daß viele der im Januar und Februar gefundenen Rauhhautfledermäuse geschwächt oder gar tot waren, unterstreicht diese Ansicht. Die Frage, auf welchem Weg die Tiere in die Sommerquartiere zurückkehren, bleibt somit weiterhin offen. Das bei Basel festgestellte Fundmaxımum im März kann durchaus mit einem Rückzug in Verbindung stehen, doch ist das Datenmaterial noch zu spärlich, um weitere Überlegungen anzustellen. GEBHARD (briefl. Mitt.) glaubt an einen geradlinigen, schnellstmöglichen Rückzug ın die Sommerquartiere. In diesem Zusammenhang ist die Mitteilung einer aus 72 Tieren bestehenden Männchengesellschaft von Pipistrellus nathusii in der deutschen Exklave Büsingen (KreisWaldshut) von Bedeutung. Jahr für Jahr kamen die Tiere regelmäßig ım April an und waren gegen Anfang Mai wieder verschwunden. Bei 208 H. V. Bastian einer durchschnittlichen Reisegeschwindigkeit von etwa 50 km/h (Kock und SCHWARTING 1987) würden die Tiere „termingerecht“ in den osteuropäischen Sommergebieten an- kommen. Die Rauhhautfledermäuse sind wahrscheinlich Saisonwanderer, die das Gebiet der Bundesrepublik überfliegen oder an wenigen besonders begünstigten Stellen überwintern. Die Art Pıpistrellus nathusu fällt demnach auch unter die Bonner Konvention vom 23. Juni 1979 über das Übereinkommen zur Erhaltung wandernder wildlebender Tierarten (Erz 1981). In die rote Liste der gefährdeten Tiere und Pflanzen ist sie unter die Kategorie II (Gefährdete Durchzügler, Überwinterer, Übersommerer, Gäste usw.) einzuordnen (BLAB et al. 1984). Danksagung Für die kritische Durchsicht des Manuskriptes danke ich den Herren Prof. Dr. E. Kurzer, Tübingen, Dr. H. Roer, Bonn, und ]J. GEBHARD, Basel, sowie Frl. Anıta v. Buch, Tübingen. Zudem bin ich noch einer Vielzahl von Fledermausforschern dankbar, die mir bereitwillig ihre Daten zur Verfügung gestellt haben. Zusammenfassung Anhand von Literaturdaten wird das Vorkommen der Rauhhautfledermaus in Baden-Württemberg beschrieben. Es wurden insgesamt 508 Funde ausgewertet. Dabei ergaben sich Fundhäufungen im Herbst und im Winter (Abb. 2). Während das Hauptmaximum im September mit einer Herbstwande- rung in Verbindung gebracht werden kann, ist das Nebenmaximum im Januar eher auf Meldungen von Winterverlusten zurückzuführen. Es wird ein Zugweg von Sommerquartieren in der DDR oder Polen durch das Neckartal zum Bodensee und von dort aus weiter durch das Schweizer Mittelland oder die Burgundische Pforte in die Winterquartiere erwogen (Abb. 5). Offen bleibt die Frage, wann und auf welchem Weg die Tiere in die Soemmerquartiere zurückkehren. Literatur AELLEN, V. (1961): La Baguement des Chauves-souris au Col de Bretolet (Valais). Arch. Sci. 14, 365-392. BAUER, K.; WIRTH, J. (1979): Die Rauhhautfledermaus (Pıpistrellus nathusii Keyserling und Blasius, 1839) (Chiroptera, Vespertilionidae) in Osterreich. Ann. Naturhistor. Mus. Wien 82, 373-385. BLag, J.; Nowak, E.; Sukopr, H. (1984): Rote Liste der gefährdeten Tiere und Pflanzen in der Bundesrepublik Deutschland. Greven: Kilda. CrauDg, C. (1976): Funde von Rauhhautfledermäusen, Pipistrellus nathusii in Zürich und Umge- bung. Myotis 14, 30-36. DIETERICH, J. (1982): Vergleichende Beobachtungen über den Fledermausbesatz in verschiedenen Nistgeräten nach Untersuchungen in Schleswig-Holstein. Myotis 20, 38—44. Erz, W. (Hg.) (1981): Schutz wandernder Tierarten. Greven: Kilda. FELTEN, H.; KLEMMER, K. (1960): Fledermäuse des Rhein-Main-Lahn-Gebietes. Bonn. zool. Beitr. Sonderheft 11, 166-188. GEBHARD, J. (1983): Die Fledermäuse in der Region Basel (Mammalıa: Chiroptera). Verhandl. Naturf. Ges. Basel. 94, 1-42. HaEnsErL, J. (1967): Notizen über 1963-1966 insbesondere in Berlin aufgefundene Fledermäuse. Milu 2, 313322. HanAaK, V.; GAISLER, J. (1976): Pipistrellus nathusi (Keyserling et Blasius, 1839) (Chiroptera: Vespertilionidae) in Czechoslovakia. Vest. Cs. spol. zool. 40, 7-23. HEISE, G. (1982): Zu Vorkommen, Biologie und Okologie der Rauhhautfledermaus (Pipistrellus nathusii) in der Umgebung von Prenzlau (Uckermark), Bezirk Neubrandenburg. Nyctalus NF1, 281-300. HızsscH, H. (1983): Faunistische Kartierung der Fledermäuse in der DDR. Teil 1. Nyctalus NF1, 489-503. IsseL, B.; IsseL, W.; MASTALLER, M. (1977): Zur Verbreitung und Lebensweise der Fledermäuse in Bayern. Myotis 15, 19-97. KLAWITTER, J. (1974): Zum Vorkommen von Pipistrellus nathusü ın Westberlin. Myotis 12, 44—45. Kock, D. (1981): Rauhhautfledermäuse im Rhein-Main-Gebiet. Natur und Museum 111, 10-24. Kock, D.; SCHWARTING, H. (1987): Eine Rauhhautfledermaus aus Schweden in einer Population des Rhein-Main-Gebietes. Natur und Museum 117, 20-29. KULZER, E. (1986): Winterschlaf von Abendseglern (Nyctalus noctula) in künstlichen Quartieren - ein Beitrag zum Fledermausschutz. Protokoll der 7. Tagung der AG-Fledermausschutz, Tübingen. Vorkommen und Zug der Rauhhautfledermaus in Baden-Württemberg 209 KULZER, E.; BasTıan, H. V.; FIEDLER, M. (1987): Die Fledermäuse in Baden-Württemberg. Beih. Veröff. Naturschutz Landschaftspflege Bad.-Württ., 50. MÜLLER, A.; WIDMER, M. (1983): Bestand und Verbreitung der Fledermäuse des Kantons Schaffhau- sen und Umgebung. Abschlußbericht, 1-82. RoEr, H. (1971): Weitere Ergebnisse und Aufgaben der Fledermausberingung in Europa. Decheniana Beih. 18, 121-144. — (1974): Die Rauhhautfledermaus (Pipistrellus nathusiu) in Mitteleuropa. Myotis 11, 18-27. — (1976): Weitere Nachweise der Rauhhautfledermaus (Pipistrellus nathusü). Myotis 13, 65-67. SCHMIDT, A. (1977): Ergebnisse mehrjähriger Kontrollen von Fledermäusen im Bezirk Frankfurt (Oder). Naturschutzarb. in Berlin u. Brandenburg 13, 42-51. — (1978): Zum Geschlechtsdimorphismus der Rauhhautfledermaus (Pipistrellus nathusii) nach Fun- den ım Bezirk Frankfurt/O. Nyctalus NFl, 41-46. SCHOBER, W. (1960): Zur Kenntnis mitteldeutscher Fledermäuse. Bonn. zool. Beitr. Beih. 18, 105-111. STRELKOV, P. P. (1969): Migratory and stationary bats (Chiroptera) of the European part of the Soviet Union. Ac. Zool. Cracov. 14, 393439. Sturz, H. P.; HAFFner, M. (1985): Geschlechtsspezifische saisonale Anwesenheit einiger mitteleuro- päischer Fledermausarten in der Zentral- und Ostschweiz. Mitt. Naturf. Ges. 32, 1-5. VIERHAUS, H.; BüLow, B. von (1978): Zwei neue Nachweise der Rauhhautfledermaus Pipistrellus nathusü (Keyserling und Blasıus, 1839) aus Westfalen. Natur und Heimat 38, 65-70. WIJNGAARDEN, A. VAN; LAAR, V. van; TROMMEL, M. D.M. (1971): De verspreiding van de uederlandse zoogdieren (Mammalia: Insectivora, Chiroptera, Rodentia, Lagomorpha, Carnivora, Pinnipedia, Artiodactyla. Lutra 13, 2—41. Anschrift des Verfassers: Hans-VALENTIN BasTIan, Institut für Biologie III der Universität Tübin- gen, Abt. für Physiologische Okologie, Auf der Morgenstelle 28, D-7400 Tübingen Z. Säugetierkunde 53 (1988) 210-224 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 The anatomy of the stomach and caecum of the Samango monkey, Cercopithecus mitis erythrarchus Peters, 1852 By M.R. BruorTon and M.R. PERRIN Department of Zoology and Entomology, University of Natal, Pietermaritzburg, South Africa Receipt of Ms. 12. 03. 1987 Abstract An investigation of the gastric and caecal anatomy of Cercopithecus mitis erythrarchus revealed a simple glandular stomach and a sacculated caecum. The unilocular glandular stomach is typical of cercopithecine primates. The caecum and colon possess well-developed taeniae and haustra, and provide microhabitats for numerous symbiotic bacteria. The morphological observations are inter- preted in terms of the functional digestive process of this primarily frugivorous primate. Introduction Herbivorous mammals can be classified into two broad categories according to their digestive strategy and gut adaptations that accommodate microbial fermentation (PARRA 1978): firstly, those anımals with an enlarged foregut that ıs the main site of microbial activity and in which microbial fermentation precedes gastric intestinal digestion, or secondly, those animals with an enlarged hindgut, in which gastric intestinal digestion of the diet occurs before microbial fermentation (PARRA 1978). Many fermentative bacteria are cellulolytic, and for those anımals with a diet rich in structural carbohydrates a large fermentation chamber in the digestive tract is necessary. Old World monkeys of the subfamily Colobinae are predominantly folivorous (CHIv- ERS and Hrapık 1980), and differ from all other primates in the large size and anatomical complexity of the stomach (BAucHoP and MarTuccı 1968). Consequently, considerable literature has accumulated on the ruminant-like, microbial gastric digestion occurring in the stomach (BaucHoP and MarTuccı 1968; OHwaRrT et al. 1974; BaucHoPr 1978) and on the microbiology of the colobine stomach (BaucHoP 1971). The more typical Old World monkeys of the subfamily Cercopithecinae are predomin- antly frugivorous- omnivorous, and the stomach is a simple, smooth-walled sac (Hırı 1958). Recent investigations on volatile fatty acıd (VFA) concentrations (the breakdown products of microbial fermentation) and digesta movement in the gastrointestinal tracts of a number of Cercopithecinae (CLEMENS and PHıLLıps 1980; CLEMENS and MaLoıy 1981) have demonstrated that the stomach contains an acıdic environment unsuitable for prolonged microbial activity. From the high concentrations of VFAs recorded in the large intestine of these primates, it appears that the caecum and colon are important sites of microbial activity and VFA production. An extensive study on the socioecology and feeding ecology of the Samango monkey Cercopithecus mitis erythrarchus ıs currently in progress in Natal (Laws pers. comm.). To examine the Samango in relation to the two digestive strategies and to relate the diet to current concepts of digestion theory, a detailed knowledge of the monkey’s gastric morphology was required. A preliminary examination of the stomach and caecum of the Samango monkey was therefore initiated to determine the gross anatomy, histology and ultrastructure of the respective organs, and to correlate anatomy with diet and function. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5304-0210 9 025010 The anatomy of the stomach and caecum of the Samango monkey 24 Materials and methods One weaned juvenile female and one adult female Samango monkey were used for this examination. The shot animals were dissected in the field immediately after death and photographed to display the position of the organs in situ. The pH in the proximal and distal stomach regions, mid-region of the small intestine, caecum, and ascending and descending colon was measured using a single glass probe electrode (Model TC pH 800) placed in a small incision ın the GIT wall. Small sections of stomach and caecal tissue (5mm x 5 mm) were fixed in 3 % buffered glutaraldehyde for scanning and transmission electron microscopy. Thereafter whole organs were placed in Bouin’s fixative for 18 h. After fixation the organs were bisected and preserved in 70 % alcohol. One half of each organ was used for gross morphological analysis, the other for histological examination using standard tissue preparation techniques (Humason 1962). Haematoxylın and eosin were used as general tissue stains, while Alcian blue was used to confirm mucus (polysaccharide) secretion (HuUMAsoN 1962). For electron microscopy tissue was removed from the gut and postfixed in 1 % osmium tetroxide in 0,05 M sodium-cacodylate buffer, then dehydrated ın a graded alcohol series. Material for transmission electron microscopy (TEM) was further dehydrated in propylene oxide and embedded in Epon. Sections were cut on an LKB Ultratome 3 ultramicrotome, stained with uranyl acetate and lead citrate and viewed with a Hitachi Model HU-11E-1 transmission electron microscope. For scanning electron microscopy (SEM), material was critical-point dried, coated with gold palladıum and viewed with a Hitachi S-570 scanning electron microscope. Results pH pH readings recorded at different sites along the GIT are given in the Table. Table. pH readings along the GIT of adult and juvenile Samango monkeys Section of tract Stomach Mid small Caecum Ascending Descending Fundus Antrum intestine colon colon Adult I. : 6.31 5497. 6.25 6.47 Juvenile i Ä 6.53 5.58 6.69 6.90 Topography in situ The stomach of ©. m. erythrarchus extends laterally across the abdominal cavity and occupies approximately one quarter by volume of the cavity. The small intestine, caecum, colon and rectum fill the remainder of the cavıty (Fig. 1). The caecum is located in the lower right region of the abdominal cavity, lyıng directly against the abdominal wall (Fig. 1) and partially obscured in the distal region by the mesentery of the small intestine. The large, haustrated transverse colon is prominent, possessing distinct taenıae (Fig. 1) along its entire length. Gross morphology Stomach The stomach is a simple unilocular glandular sac (Fig. 2a), with the oesophagus entering the mid-dorsal region of the corpus (forestomach). Sıinistral to the gastro-oesophageal junction the fundus extends laterally, torming a slightly globular pouch. Dextral to the junction, the lesser curvature immediately reflects to form the pars pylorica (Fig. 2a); the antrum (hindstomach) extends from the oesophageal opening, and curves dorsally forming 212 M. R. Bruorton and M. R. Perrin DENEN Fig. 1. The disposition of the GIT within the abdomen of C. m. erythrarchus. Note the haustrated transverse colon (TC) and the taenıa (t). SI = small intestine, C = caecum, $ = stomach the pars pylorica. The only apparent difference between the stomachs of juvenile and adult monkeys was the increased folding of the stomach wall in the adult. Caecum The caecum of adult C. m. erythrarchus dısplays several haustrum-like sacculations and two prominent, laterally sıtuated taenıae which extend along the length of the colon. The ileum enters the caecum at the border between the ampulla caecı and corpus caecı (Fig. 3). Distal to this opening at the end of the ampulla ıs the caecocolic junction, from which the proxımal colon emerges (Fig. 3). Although the general form of the adult and juvenile caeca ıs similar, they differ markedly in the extent of haustration and folding of the internal mucosal surface. In the adult (Fig. 3), a deep sacculation separates the corpus caecı and ampulla; it ıs likely that digesta is directed into the corpus caeciı via a muscular valve at the ıleocaecal junction (Fig. 3). The extent of the internal folding is similar in the caecum and proximal colon. The caecum of juveniles (Fig. 4), in comparison, is a large unilocular sac with no distinct separation between the corpus caecı and ampulla caecı. Histology Stomach A typically mammalian gastric tissue plan (WEıss and GrEEP 1977; Ham and CoRMACK 1979) is present (Fig. 5). The muscularıs externa is generally thicker in the corpus than in the antrum, where the circular muscle layer (stratum circulare) is approximately three times thicker than the longitudinal layer (stratum longitudinale). The anatomy of the stomach and caecum of the Samango monkey 28 2b PPC. D ze, Pyloric x Cardiac =— m. Fundic Glandular tissue Muscularis externa Oesophaqus nn, il (hH Duodenal epithelium Fig. 2a. A photograph of a bisected stomach of C. m. erythrarchus (juvenile). F = fundus, C = corpus, Pp = pars pylorica, O = oesophagus, Ps = pyloric sphincter, a = antrum. — b: Semi-diagrammatic drawing of a bisected stomach showing the regions of glandular tissue A submucosa of loose connective tissue is of constant thickness in the corpus and antrum (Fig. 5). The muscularıs mucosa is continuous with that of the oesophagus, and lamina propria extends between the glandular tissue. The mucosal lining of the stomach is histologically divisible into cardıac, fundic and pyloric regions. The cardiac tissue in the stomach extends 1-1,5 cm radially from the cardiac orifice (Fig. 2b). The short, branched cardıac glands possess wide foveolae with closely packed mucous neck cells (Fig. 6). These cells continue for some distance into the tubules of the glands. Parietal (oxyntic) and chief (or peptic or zymogen) cells are present 214 M. R. Bruorton and M. R. Perrin Fig. 3. A photograph of the bisected caecum of an adult C. m. erythrarchus, showing the deep haustration between corpus caecı (Cc) and ampulla caecı (Ac), and the muscular valve (Mv) at the ıleocaecal junction. I = ıleum Fig. 4. A photograph of the bisected caecum of a juvenile C. m. erythrarchus. Cc = corpus caeci, Ac = ampulla caeci, PC = proximal colon (Fig. 6), but their relative scarcity and the large numbers of closely packed mucous cells is characteristic of cardıiac glands. In the fundic region the glands are long and tubular with narrow foveolae and extend through the mucosa to the muscularis mucosa (Fig. 7). A histological stratification typical for mammalian fundic glands is evident, i.e., cuboidal neck cells, followed by mid-region mucous neck cells, and chief cells at the base with parietal cells scattered throughout the gland, but predominantly in the mid-gland regions (Fig. 7). In the pyloric region parietal and chief cells are absent. Typically only one cell type ıs evident, a mucus-secreting cell similar to the mucous neck cells of the fundic region (Fig. 8). Pyloric glands are highly coiled and possess wide foveolae (Fig. 8). The anatomy of the stomach and caecum of the Samango monkey 215 % r r % BE E U EEE Fıg. 5. A typical mammalian gastric tissue plan was observed. G = glandular tissue, Mm = muscularis mucosa, Sb = submucosa, Cm = circular muscle layer, Lm = longitudinal muscle layer. — Fig. 6. A section of the cardıac glands showing mucoid neck cells (Mc), parietal cells (P) and chief cells (C). F = foveolum of gland. — Fig. 7. A photograph of a section of fundic glands showing the long, tubular nature of the glands. M = mucus, P = region of parietal cells, C = region of chief cells. — Fig. 8. A photograph of a pyloric gland. F = foveolum of gland, Mc = mucus-secreting cells 216 M. R. Bruorton and M. R. Perrin Caecum The caecum also possesses a typical mammalian alimentary tissue plan (Fig. 9). The muscularıs externa ıs continuous with that of the ileum and the proximal colon; the submucosa of loose connective tissue maintains a constant thickness throughout the caecum (Fig. 9). The muscularıs mucosae is continuous with that of the ileum and colon. A lamina propria of fine connective tissue and elastin fibres is present above the muscularis mucosae and extends between the glandular tissue (Fig. 9). The surface epithelium consists of predominantly columnar absorptive cells, and numerous crypts with wide openings that extend along the length of the caecum. These glands contain numerous columnar mucous goblet cells (Fig. 10). Fig. 9. A photograph showing the typical mammalian alımentary tissue plan of the caecum. G = glandular region, Sb = submucosa, Cm = circular muscle layer, Lm = longitudinal muscle layer. - Fig. 10. A photograph of two caecal crypts, showing the abundance of goblet cells (G) and the lamina propria (Lp) between the crypts Histologically, no differences are apparent between the ampulla and corpus caecı, or between juvenile and adult caeca. Electron microscopy Stomach Material examined by SEM was limited to fundic and pyloric tissue. The pyloric region of the stomach is characterised by deep infoldings (Fig. 11). The pylorie glands open into these recesses and the columnar epithelial cells show a typical “cobblestone” surface appearance (Fig. 11). Numerous sessile filamentous bacteria ın palisade configuration were present in the adult fundus (Fig. 12), while ın the juvenile few isolated pockets of shorter bacteria were found (Fig. 13). No bacteria were present in the pars pylorica of either monkey. The anatomy of the stomach and caecum of the Samango monkey 219, Caecum SEM observations support the light-microscopic findings. The surfaces of both the corpus caecı and ampulla caecı contain numerous crypts and a dense covering of long, threadlike microvilli (Figs. 14 and 15). TEM sections through the caecal crypts show that the crypts are surrounded by tall columnar epithelial and numerous goblet cells (Fig. 16). The nuclei of the goblet cells are basally situated and numerous supranuclear mitochondria are present. Large numbers of cocco-bacılli and cocci occur, apparently unattached, among the a % Gi 4 Fıg. 11. A scannıing electron micrograph of two pyloric glands, showing the foveola of the glands (F) and the typical cobblestone appearance of epithelial cell surfaces (Ec). - Fig. 12. Bacilli in the fundus of adult C. m. erythrarchus. — Fig. 13. Cocco-bacilli in the fundus of juvenile C. m. erythrarchus. — Fig. 14. A micrograph of the caecal epithelial surface showing the dense layer of microvilli (Mv) and the caecal crypts (Ce) 218 M. R. Bruorton and M. R. Perrin Fig. 15. A high-power micrograph of caecal microvilli. - Fig. 16. A transmission electron micrograph of a section through a caecal crypt, showing bacteria (B) in the crypt lumen and the mucoid body (M) of a goblet cell. Mt = mitochondria. — Fig. 17. A micrograph showing bacteria (B) on the caecal surface. Mv = microvilli. - Fig. 18. A micrograph showing the close association of a bacterıum (B) with the microvilli (Mv). G = glycocalyx The anatomy of the stomach and caecum of the Samango monkey 2) mucus and food particles on the caecal surface (Fig. 17) and also penetrate into the caecal crypts. Under TEM the close association of the bacteria in the crypt lumen with the microvillar border is striking (Fig. 18). Many of the bacteria are lodged between the microvilli and appear to indent the apical plasma membrane of the cells (Fig. 19), but no bacterıal penetration of cells ıs evident. | Fig. 19. Bacteria in the lumen of a caecal crypt. Some bacteria indent the plasma membrane (arrows) Discussion The food sources of primates can be classified into three major groups, depending on their structure and biochemical composition, and resulting digestibility (CHivers and Hranık 1980). These are fruits, leaves and anımal matter. “Fruits” include unripe (e.g. flowers) and ripe (fleshy) parts, seeds and tubers, i.e., mostly the reproductive parts of plants, which are foods containing short-chain sugars that are hydrolyzed rapıdly for absorption and immediate use. “Leaves” include young and mature leaves, grasses, stems and also barks and gums, ı.e., the vegetative parts of plants, which are foods usually containing protein and long-chain sugars that require fermentation in an enlarged stomach or large intestine. “Anımal matter” includes invertebrates and small vertebrates, which provide sources of protein and fat that are easily digested and, therefore, require a relatively short and simple gut (CHivErs and Hrapık 1980). C. m. erythrarchus has the following percentage volumetric dietary composition for 713 plant food items scored ın the Cape Vidal dune forest: fruits 69.2 %; leaves 23.6 %; anımal matter 4.8 %; other 1.6 % (Lawes 1986, unpubl. data). Thus, the Samango is a typical frugivore (CHivers and HLADIK 1980), subsisting predominantly on fruits and supplementing the diet wıth varyıng amounts of leaves and insects. Fruits are easily digested and absorbed, while leaves require a fermentation process if a maximal energy return from the foliar component of the diet is to be achieved (Mırron 1981). From a knowledge of the diet and the gastric anatomy of the stomach and caecum of C. m. erythrarchns, it is possible to speculate on the digestive and foraging strategy used by the anımal and whether a fermentation process is likely in the gut. Stomach The stomach of C. m. erythrarchus ıs a simple glandular unilocular sac, typical of the subfamily Cercopithecinae (Hırı 1958; LANGER 1985). In the stomach three main glandu- lar areas of the mucosa, the cardiac, fundic and pyloric regions, are ıdentified. The gastric 220 M. R. Bruorton and M. R. Perrin secretions of proteolytic enzymes, hydrochloric acid and mucus are associated with these three areas of the stomach (LAnGER 1985). The well-developed gastric muscle layers and numerous rugae of the adult may facilitate mechanical digestion (the mixing, grinding and soaking of digesta) and provide a greater surface area for secretion, respectively, possibly as a result of the greater nutritoinal demand of the adult monkey. The primarily frugivorous diet of the Samango, however, does not require much mechanical breakdown in the stomach. The last premolars of frugivores tend to become molariform to provide additional crushing and grinding surfaces (MAIER 1984), while the cusps and edges of the molars of the Samango are bunodont (low and blunt). Thus, foods can be masticated to such a degree as to allow the maximum surface area for digestion, and the function of the rugae in the adult may simply be to provide a greater surface area for the production and release of gastric secretions. Bacteria were present in both adult and juvenile Samango stomachs. It is highly unlikely, however, that the presence of small numbers of bacteria in the unilocular (ancestral, unspecialized; CARLETON 1973) stomach is indicative of a fermentation process (Mappvock 1981). Where microbial fermentation occurs in the stomach, a voluminous fermentation vat is differentiated, holding food for fermentation and slowing the transit of digesta through the stomach (LAnGer 1984b). The wholly glandular unilocular stomach of C. m. erythrarchus ıs obviously not suited to this function, since there is no physical or physiological separation of digestive processes in the fore and hindstomach. This is supported by the variable but high pH, as a decline to alow pH would destroy any bacteria present. The stomach of ©. m. erythrarchus, theretore, likely functions in the initial digestion of the high energy, easily digestible soluble carbohydrate and protein component of the diet. It is highly unlikely that any bacterial cellulolytic fermentation occurs, and although some initial gastric digestion of the foliar component may occur in the stomach, fermentative breakdown of leaves must surely occur elsewhere in the gut. Caecum No multicellular anımal ıs known to synthesise cellulase and in consequence the anımal is deprived not only of the nutrient value of the cellulose itself, but also of the digestible cell contents bounded by the cellulose wall (Janıs 1976). Therefore a great advantage can be gained by an anımal that can include fibrous components in ıts diet by entering into a symbiotic association with cellulase-producing bacteria, but it must provide a fermentation chamber within the digestive tract where these bacteria can digest the cellulose (Jans 1976). The importance of the caecum and colon in fermentation processes has been determined for many herbivorous mammals (CURRIER et al. 1960; JoHNsoN and McBEE 1967; Yanc et al. 1968; McKENZIE 1978; SnıpEs 1978, 1979; CLEMENS and PHıLLıps 1980; CLEMENS and Maroıy 1981; Van Hoven et al. 1981; Snıpes 1981, 1982a, b, 1984a, b). It is possible that the caecum of C. m. erythrarchus functions as a fermentation chamber for the breakdown of the protein and fibre-rich leaf component of the diet, and evidence for this speculation is delivered in the investigation. The major products of the fermentation of carbohydrates are usually the volatile fatty acıds (VFAs), acetic, propionic and butyric (and methane). Fermentative activity ıs usually measured by determining VFA concentration in a portion of the alımentary tract (PARRA 1978). The presence of these organic acids along the digestive tract of non-ruminant herbivores has been demonstrated in vivo repeatedly (RERAT 1978); their concentrations are very large and reach a maximum at the level of the caecum and colon. Studies on four species of African Cercopithecinae have confirmed this distribution of VFAs along the digestive tract (CLEMENS and Prırrıps 1980; CLemeEns and Maroıy 1981). The concen- The anatomy of the stomach and caecum of the Samango monkey DD tration of organic acids in the stomach is low, with lactic acid representing over half the total amount. Their concentration remains low in the small intestine but increases abruptly in the caecum and proximal half of the colon, where the passage of digesta is probably slowed down greatly by the taenıae and haustra (CLEMENS and PHırLıps 1980; CLEMENS and Marory 1981). The high concentrations of VFAs are strong evidence for an extensive microbial fermentation in the caecum and colon of these four primate species. Microbial fermantation within the digestive tract of primates is a relatively new concept (CLEMENS and PHırrıps 1980). With the exception of the Colobinae with their complex foreguts (BaucHoP and MarTuccı 1968; OHwarI et al. 1974), ıt was believed until recently that primates were generally incapable of fermenting and digesting complex carbohydrates. However, the concentration of organic acıds produced in the caecum and colon of the baboon, Sykes monkey, Vervet monkey and bushbaby (CLEmEns and PrirLıps 1980; CLEMENS and Maroıy 1981) are comparable to those observed in the foregut of ruminants and the hindguts of dogs, pigs and ponies (SwENSoN 1982). Considerable numbers of bacteria (representing a large biomass) are present ın the caecum of C. m. erythrarchus and the close association of many bacteria with the epithelium is striking. In the Koala (Phascolarctos cinerens) the close association of bacteria with the epithelium aids absorption by epithelial cells of the products of microbial fermentation (MckENZIE 1978). In the Samango, an actual penetration of bacteria into the epithelium is not evident, as was characteristic in the Koala. Rather, the adherence of the bacteria to the epithelium is in the form of an intermicrovillar location of the bacterıa. Together with the large numbers of bacteria found among food particles, the presence of the adhering bacteria suggests that the caecum and microflora function symbiotically as a fermentation vat. In C. m. erythrarchus the pH values along sections of the gastrointestinal tract are very similar to those recorded for two congeneric primate species, C. mitis and C. pygerythrus (CLEMENS and PHırLıps 1980; CLEMENS and MaLo1y 1981). High concentrations of VFAs were present in the caeca of these two species, leading to a lowering of the pH. Due to the similar pH values along the GIT ot all three species and the fact that high concentrations of VFA’s have been observed in C. mitis and C. pygerythrus, ıt seems likely that the low pH in the caecum of C. m. erythrarchus ıs a result of the presence of hıgh concentrations of VFAs. These acıds are formed as breakdown products of microbial fermentation; this evidence, together with the observed large biomass of bacteria, also suggests that the caecum functions as a fermentation chamber. These aspects of physiology are to be investigated in C. m. erythrarchus. The close proximity of ileocaecal and caecocolıical orifices has been observed in previous studies (Snıpes 1981, 1982a, b). This is interpreted as providing the possibility of a rapid transport of intestinal contents directly from ıleum to colon, thereby by-passing the greater portion of the caecum (Snıpes 1982b). This probably occurs with protein-rich, fibre-poor foodstuffs. Protein-poor and fibre-rich foodstuffs, however, require long-term fermentation activity for the necessary breakdown of cellulose, which would occur in the body of the caecum under optimal fermentation conditions (Snıpes 1982b). The presence of extensive musculature in the ileocaecal junction of C. m. erythrarchus, which appears to act as a valve/sphincter, could aid this directional flow, thereby enabling optimal fermenta- tion conditions to persist in the caecum. The possible significance of caecal and colonic taenıae and resulting huastra has been discussed by GABELLA (1982 op. cit. SnıpEs 1982b, 1984a). He implies that the condensa- tion of the longitudinal muscle into narrow bands (taeniae) may represent a specialization allowing mixing and movement of contents, as would be favourable for a “fermentation- vat” function of the caecum. Haustra are interpreted as advantageous for the fermentation process, acting as areas for slowing the rate of passage of digesta and thus allowing the process of fermentation to occur in a more suitable environment (LANGER 1979; CLEMENS 222 M. R. Bruorton and M. R. Perrin and PhırLıps 1980; LANGER 1984a). The possession of well developed taeniae and haustra along the caecum and colon again strongly suggests active fermentation in these regions. The striking structural differences between adult and juvenile caeca may be attributable to the fermentation process. The adult possesses a relatively small caecum with numerous folds and a large, haustrated colon, while the juvenile has a large caecum and a relatively small colon. In the adult, although caecal fermentation almost certainly occurs, colonic fermentation may be more important to the monkey. In the juvenile the colon has not developed to adult size and the caecum is proportionately larger and possibly more important in the fermentation process. JOHNSON and McBeE (1967) state that caecal fermentation in the porcupine (Erethizon dorsatum epixanthrum) appears to be of increa- sing importance ın larger, older anımals, and that this could be offset by younger animals having a proportionately larger caecum (JoHnson and McBee 1967). Four major facts exemplify the importance of hindgut fermentation as an energy source in C. m. erythrarchus: 1. The presence of taenıae and haustra ın the caecum and colon. 2. The resulting slow rate of passage of digesta (GABELLA 1981 op. cit LANGER 1979; SNIPES 1984a). 3. The presence of numerous bacteria interspersed within the microvilli. 4. The high VFA concentrations in congeneric species with similar anatomies and pH values. However, the extent of the fermentation process in the caecum and colon of ©. m. erythrarchus ıs not known at present and is the subject of further investigation. Conclusion It is possible to interpret the morphology of the stomach and caecum of C. m. erythrarchus in terms of the dietary requirements of the anımal. The Samango eats primarily fruit (Lawes 1986 unpub. data), which ıs generally rich ın nonstructural carbohydrates but low in protein. The simple, glandular stomach provides copious secretions of acıds and enzymes necessary for the breakdown of the high quality and easily digestible fruit component, with consequent high utilisation of the nonstructural, soluble carbohydrates. However, the Samango supplements its primarily frugivorous diet with a considerable amount of foliar material (23.6 % of dietary composition). Leaves are high ın protein and structural carbohydrates, and to maximize energy returns from leaves requires slow passage rates of food for microbial fermentation to occur. The convoluted caecum and capacıous colon of the Samango both possess taenıae and haustra, which slow down the passage rate of food and suggest that an active fermentation process occurs. This ensures maximal energy returns from the foliar component of the diet and from undigested fruit pectins. The caecal and colonic fermentation would be particularly important in mid- winter when food quality and abundance is low, and Samango’s experience a “crunch” period with a relatively nutrient-poor diet. Acknowledgements We would like to thank Mr. M.]J. Lawes and Mr. R. FErGuson for their help in collecting the specimens, and Mr. Lawes for allowing us free access to data collected from his Cape Vidal study area. Our thanks are extended to the staff of the Pietermaritzburg University Electron Microscope Unit for their technical assistance and to Mr. A. H. Mappvock for constructive and stimulating discussions. The anatomy of the stomach and caecum of the Samango monkey 223 Zusammenfassung Die Anatomie von Magen und Blinddarm der Diadem-Meerkatze Cercopithecus mitis erythrarchus Peters, 1852 Eine Untersuchung der Magen- und Blinddarm-Anatomie von Cercopithecus mitis erythrarchus zeigte einen einfachen Drüsenmagen. Der einhöhlige Magen ist für die cercopithecinen Primaten typisch. Blinddarm und Dickdarm sind mit gut entwickelten Tänıen und Haustren ausgestattet, die einen Lebensraum für viele symbiontische Bakterien darstellen. Die morphologischen Figentümlichkeiten werden funktionell in Zusammenhang mit den Verdauungsprozessen dieses hauptsächlich fruchtfres- senden Primaten interpretiert. References BAucHoP, T.; MAarTuccı, R. W. (1968): Ruminant-like digestion of the Langur monkey. Science 168, 698-700. BAucHop, T. (1971): Stomach microbiology of primates. Ann. Rev. Microbiol. 25, 429-436. — (1978): Digestion of leaves in vertebrate arboreal folivores. In: The ecology of arboreal folivores. Ed. by G. G. MontGoMERy. Washington D.C.: Smithsonian Inst. Press. CARLETON, M. D. (1973): A survey of gross stomach morphology in New World Cricetinae (Rodentia, Muroidea) with comments on functional interpretations. Mus. Zool. Univ. Michigan 146, 1-42. CHivErs, D. J.; Hıapık, C. M. (1980): Morphology of the gastrointestinal tract in primates: comparisons with other anımals in relation to diet. J. Morph. 166, 337-386. CLEMENS, E. T.; PhHiırLıps, B. (1980): Organic acıd production and digesta movement in the gastrointestinal tract of the baboon and Sykes monkey. Comp. Biochem. Physiol. 66 (A), 529-532. CLEMENS, E. T.; MAaro1y, G. M. ©. (1981): Organic acıd concentrations and digesta movement in the gastrointestinal tract of the bushbaby (Galago crassicandatus) and Vervet monkey (Cercopithecns pygerythrus). J. Zool., London, 193, 487-497. CURRIER, A.; Kırtts; Cowan, M. C. T. (1960): Cellulose digestion in the beaver (Castor canadensis). Can. ]J. Zool. 38, 1109-1116. Ham, A. W.; Cormack, D. H. (1979): Histology. 8th. Ed. Toronto: J. B. Lippincott Company. Hırı, W. C. ©. (1958): Pharynx, oesophagus, stomach, small and large intestine. Form and position. Primatologıa 3, 139-207. Humason, G.L. (1962): Animal tissue techniques. W. H. Freeman and Company. Hume, I. D.; WARNER, A. C. 1. (1980): Evolution of microbial digestion in mammals. In: Digestive physiology and metabolism in ruminants. Ed. by Y. RuckEsuscH and P. THıvanD. Westport, Connecticut: AVI Publishing Co. Janıs, C. (1976): The evolutionary strategy of the Equidae and the origins of rumen and caecal digestion. Evolution 30, 757-774. Jornson, J. L.; McBer, R. H. (1967): The porcupine caecal fermentation. J. Nutrition 91, 540-546. LANGER, P. (1979): Functional gastric anatomy of macropod marsupials. Ann. Rech. Vet. 10, 476-479. — (1984b): Comparative anatomy of the stomach in mammalıan herbivores. Quart. J. Exper. Physiol. 69, 615-625. — (1984a): Anatomical and nutritional adaptations in wild herbivores. In: Herbivore nutrition ın the subtropics and tropics. Ed. by F. C. M. GircHRIST and R. ]. Mackie. Craighall, South Africa: Science Press. — (1985): The mammalian stomach: structure, diversity and nomenclature. Acta. Zool. Fennica 170, 321.92: Mappock, A.H. (1981): The gastric morphology of the white-tailed rat Mystromys albicaudatus (A. Smith 1834) and preliminary investigations of its digestive processes. M. Sc Thesis, Rhodes Univ., Grahamstown, South Africa. MAIER, W. (1984): Tooth morphology and dietary specialization. In: Food acquisition and processing in primates. Ed. by D. J. CHivers, B. A. Woop and A. BıLsBOROUGH. New York: Plenum Press. McKEnNZIE, R. A. (1978): The caecum of the Koala, Phascolarctos cinereus: light, scanning and transmission electron microscopic observations on its epithelium and flora. Aust. J. Zool. 26, 249-256. Mırrton, K. (1981): Food choice and digestive strategies of two sympatric primate species. Amer. Nat. 117, 496-505. OHwaRI, K.; HUNGATE; LOTTER; HOFMANN; Marory (1974): Stomach fermentation in Fast African Colobus monkeys in their natural state. Appl. Microbiol. 27 (4), 713-723. PARRA, R. (1978): Comparison of foregut and hindgut fermentation in herbivores. In: The ecology of arboreal folıvores. Ed. by G. G. MONTGOMERy. Washington D.C.: Smithsonian Press. 224 M. R. Bruorton and M. R. Perrin Rerar, A. (1978): Digestion and absorption of carbohydrates and nitrogenous matters in the hindgut of the omnivorous nonruminant anımal. J. Anım. Scı. 46 (6), 1808-1837. Snıpes, R. L. (1978): Anatomy of the rabbit caecum. Anat. Embryol. 155, 57-80. — (1979): Anatomy of the caecum of the Vole, Microtus agrestis. Anat. Embryol. 157, 181-203. — (1981): Anatomy of the caecum of the laboratory mouse and rat. Anat. Embryol. 162, 455474. — (1982a): Anatomy of the caecum of the Gerbil Meriones ungnicnlatus (Mammalia, Rodentia, Cricentidae). Zoomorph 100, 189-202. — (1982b): Anatomy of the Guinea-pig caecum. Anat. Embryol. 165, 97-111. — (1984a): Anatomy of the caecum of the West Indian manatee, Trichechus manatus (Mammalıa, Sirenia). Zoomorph 104, 67-78. — (1984b): Anatomy of the caecum of the cat. Anat. Embryol. 170, 177-185. SwEnson, M. J. (1982): Duke’s physiology of domestic anımals. 9th Ed. London: Cornstock Publishing Ass., Cornell Univ. Press. Van Hoven, W.; PRıns; LANKHURST (1981): Fermentative digestion in the Afrıcan elephant. $. Afr. ]. Wildl. Res. 11, 78-86. Weıss, L.; GREEP, R. ©. (1977): Histology. 4th Ed. New York: McGraw-Hill Book Company. Yang, M. G.; MANOHARAN; Young (1969): Influence and degradation of dietary cellulose ın the caecum of rats. J. Nutrition 97, 260-264. Authors’ address: MiCHAEL R. BRUORTON and Prof. Dr. MıcHAEL R. PERRIN, Department of Zoology and Entomology, University of Natal, P. ©. Box 375, Pietermaritzburg 3200, Natal, South Africa Z. Säugetierkunde 53 (1988) 225-234 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Structure de l’espace et retour au nid chez la gerbille de Mongolie (Meriones unguiculatus) Par M. LE BErRE et L. LE GUELTE Laboratoire d’Ethologie, Universite Claude Bernard, Lyon Reception du Ms. 5. 3. 1987 Abstract Space utilization and homing in the Mongolan gerbil (Meriones unguiculatus) The retrieving performance of two groups of ten different Mongolian gerbil adult females and the spontaneous homing performance of the pups were tested under two different conditions of environment from parturition to weaning. The first environment (long cages) was a very long cage (120 x 15 cm), with a closed nesting box. The second (standard cages) was a traditional square rearing- cage (40 x 40 cm), with an open nest. Latencies of first mother-pup contact during the tests under the two conditions did not differ significantly. Retrieval rate was higher from Day 06 to Day 18 in the long cages than in the standard cages. High retrieval rate was noted over a longer period of time in the long cages than in the standard cages. The reverse was the case for spontaneous homing, which made its appearance on Day 10 in the long cages and on Day 02 in the standard cages. These behavioural differences were related to the difference in structure of the available space which might be classified as a “breeding site” and an “extra-nest environment”. These experiments show the adaptability of the adult female’s behaviour to particular environmental design, which, in natural environments, may be of considerable value for the survival of the pups. Introduction Chez de nombreuses especes de rongeurs, le succes reproducteur est directement lie aux relations comportementales qui s’etablissent entre parents et enfants, en particulier durant la periode d’allaitement, en raison du caractere nidicole de la plupart de ces jeunes. Le ramassage des jeunes (retrieving des auteurs anglo-saxons) est une de ces relations comportementales dont la manifestation est fortement dependante de l’espece Etudiee. Le comportement de ramassage a Et& decrit chez la plupart des rongeurs nidicoles, ä l’excep- tion des Dipodidae (genre Jaculus), chez qui cette actıvite ne semble pas exister (BERNARD 1969). L’activite de ramassage s’observe lorsque des jeunes ont quitte le nıd ou en ont e£te retires experimentalement. Le jeune est ramen& au nid par l’un ou l’autre des parents qui le tient A la gueule. Au fur et ä mesure du developpement de ses capacıtes locomotrices, le jeune retourne lui-m&me au nıd. Cette activit peut Etre interpretee comme une relation comportementale manifeste entre les parents et leurs jeunes. Il peut aussi signifier que l’espace accessible ä ces anımaux, m&me dans les conditions de la captivite, n’est pas homogene et isotrope mais presente des qualites locales favorisant la pratique d’un certain type d’activite. Le comportement de ramassage des jeunes peut alors Etre envisage comme l’expression d’une structuration, par les anımaux, de l’espace disponible pour leurs comportements. Dans cette optique de structuration spatiale non aleatoıre des activites, l’espace peut Etre divise en une zone «favorable a l’elevage des jeunes» (c’est ä dire, le nıd sensu stricto) et une zone «non favorable ä l’elevage des jeunes», cette derniere pouvant £tre exposee a des conditions climatiques nefastes ou & l’influence de predateurs. Les jeunes en bas äge, n’ayant pas de regulation thermique efficace, se refroidissent des qu’ils sont hors du nid. Cet inconfort thermique se traduit par des @missions d’ultra-sons chez M. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5304-0225 $ 02.50/0 226 M. Le Berre et L. Le Guelte ungniculatus (DE GHETT 1974). Dans ce cas, cette division est attest&e par le fait qu’au lieu de se coucher sur les jeunes pour les rechauffer au contact de leur corps, les parents les ramenent systematiquement dans le nid. Chez certaines especes, comme Mus musculus (NOIROT 1969), Baiomys (BrLaır 1941), Peromyscus (DupLex 1974), Microtus (Hatfield 1935), le comportement de ramassage est partage par les deux parents. Chez d’autres especes, le ramassage des jeunes est davantage le fait de l’un des parents seulement. Chez Mesocricetus auratus (MARQUES et VALENSTEIN 1976), les mäles adultes effectuent plus frequemment cette actıvite que les femelles. Au contraire, chez Meriones ungniculatus, c’est presque exclusivement la femelle adulte qui effectue le ramassage des jeunes (ELWooD 1975; WARING et PERPER 1979). Cette activite ne represente chez les mäles de cette espece que 0,09 % de leur activite totale dans la periode qui s’etend de la naissance au sevrage (ELwooD 1975). Selon WarınG et PERPER (1979), la frequence de ramassage €volue de la naissance au sevrage, passant par un maximum cing Jours apres la naissance, elle decroit par la suite et s’acheve au 18Eme jour post-partum. L’evolution de ce comportement semble dependre de variations de la r&activite de la femelle jusqu’au dixieme jour post partum, et serait liee, par la suite, & l’evolution des capacites motrices et de l’aptitude des jeunes A trouver eux-m&mes le chemin du nid. Nous avons essay& d’aller au delä de cette explication, en testant l’effet de la structure du site de nidification, d’une part sur l’intensit€ du ramassage et d’autre part sur la duree d’expression de ce comportement de ramassage tout au long de la periode d’allaitement. Dans son environnement naturel, la gerbille de Mongolie eleve ses jeunes dans un terrier sombre et profond, le nid est ainsi tres nettement separe du reste de l’espace ou Evolue l’anımal. Dans le terrier, la pente de la galerie reduit les possibilites de divagation des jeunes (BANNIKOV 1954). Les cages classiques de laboratoire utilisees aussı bien par WArınG et PERPER (1979) que par nous-m@mes, sont des enceintes carr&es, de matiere plastique, exposees a la lumiere artificielle, ou le nid et le reste de la cage forment un continuum spatial. WARING et PERPER (1979) ont montre que, dans cette situation experimentale, le comportement de ramassage Evolue en relation avec l’Etat du milieu interieur de la femelle, avec la maturation et la mise en place des aptitudes locomotrices des jeunes. Nous avons envisage l’hypothese que l’expression du comportement de ramassage depend aussi de la structure de l’espace disponible pour l’activit des anımaux. Materiel et methodes Afın de tester cette hypothese, nous avons utilise deux types d’enceintes experimentales qui differaient par les variables suivantes: — La forme ‚generale: cage longue et Etroite dans un cas, cage carree dans l’autre cas. — les caracteristiques physiques de l’environnement du nid: suivant les cas, le nid £tait situe: 1. Soit dans un espace restreint susceptible de fournir une intense stimulation tactile, ou l’Eclairage etait reduit, rappelant les conditions du nid souterrain, et dans lequel l’entassement peut augmenter les stimulus oltfactifs lies au nid (longues cages). 2. Soit dans des cages bien Eclairces, avec des nids ouverts (cages standard). Dans ce dernier cas, le nid n’est pas separe physiquement du reste de l’enceinte d’elevage, alors que dans l’autre cas, une cloison separait l’espace de nidification du reste de la cage. Ainsi, la structure de l’espace imposait aux gerbilles adultes une specialisation pour la r&alisation de leurs actıvites: entretien du nid et periodes de repos dans la boite-nichoir, activite locomotrice, alımentation et prise de boisson dans le reste de la cage. Longues cages Longues cages £troites (LX1x h= 150 x 15 x 40 cm), sol en bois, paroi vitree le long d’un grand cöte, grillage sur l’un des petits cötes (fig. 1). A l’autre extr&mite, une boite-nichoir de 15 x 15cm communiquait avec la cage par un orifice de 4 cm de diamötre. Cette boite-nichoir etait munie d’une Structure de l’espace et retour au nid chez la gerbille de Mongolie 22, porte exterieure accessible a l’experimentateur. La mangeoire et l’abreuvoir £taient situes dans la partie oppossee de la cage. Le jour de la naissance, les adultes, leur portee et les materiaux constituant leur nid usuel &taient places dans la boite-nichoır. Abreuvoir Nichoir Mangeoire 15cm Fıg. 1. Longue cage (L. C.) Cages standard Il s’agissait de cages standard d’elevage (40 x 40 x 15 cm) ferm&es par un couvercle superieur de barreaux d’acier (tig. 2). Comme usuel, la nourriture et le biberon £taient inclus dans le couvercle. Le fond de la cage £tait couvert de sciure de bois. Les nids Etaient toujours construits dans un angle de la cage, avec des debris de papier. Dans les deux types de cage, les anımaux disposaient de feuilles de papier-ouate destine ä &tre dechiquete et a completer la construction du nid. #———4lcm — Litiere Nid sensu stricto Fig. 2. Cage standard (C. $.) Sujets d’experience Les couples reproducteurs utilises dans cette experience provenaient de l’elevage du laboratoire d’Ethologie. Ils etaient Eleves comme couples permanents. Dix couples differents et leurs portees ont ete utilises dans ce protocole, pour chaque type de boite. Lenombre median de jeunes par portee £tait de cing. La parite mediane des femelles Etait de deux aussi bien pour la serie des longues cages que pour les cages standard, avec quatre femelles primipares dans chaque condition. Conditions d’elevage Dans les deux types de cage, la tempe&rature ambiante £tait de 22 + 2°C. Un cycle automatique d’eclairage (14 heures de jour/10 heures de nuit) allumait la lumiere ä 7 heures du matin. La nourriture (sous torme d’aliments compresses UAR A03) ainsi que l’eau de boisson &taient disponibles en permanence ad libitum. Une fois par semaine, des graines de tournesol &taient distribuees. 228 M. Le Berre et L. Le Guelte Protocole experimental Le jour de la parturition constitue le jour JOO. Les tests sont effectu&s quotidiennement de JO1 & J19, entre 11.00 et 12.00 heures. Une cage de chaque type est testee en alternance. Juste avant le test, les parents sont enleves de la cage. Trois jeunes, pris au hasard, sont enleves du nid et places sur le plancher de la cage, ä trente centimetres du nid. La femelle adulte est alors introduite dans la cage par la partie la plus eloignee du nid. Le moment d’introduction de la femelle est le temps TO de la periode d’observation qui dure dıx minutes. Nous enregistrons les evenements suivants (ainsi que les moments et durees de ces actıvites): — premier contact entre la mere et l’un des jeunes; — premier, second et troisieme acte de ramassage (c’est a dire moment ou la femelle ayant ainsi le jeune dans la gueule commence ä le transporter; les Echecs d’une tentative de transport vers le nid ne sont pas pris en compte); — premier, second et troisieme retours spontanes des jeunes dans le nid, le cas Ech£ant, avant la fin des dix minutes; — les jeunes qui n’ont pas Et€ ramass&s ou qui n’ont pas regagn& le nıd par leur propres moyens sont consideres comme «restant hors du nid». Traitements statistiques La comparaison, entre les deux conditions, du nombre total de j jeunes ramasses, du nombre total de jeunes rentrant spontanement au nid, ou de jeunes non ramasses a Ete r&alısee a l’aide de tests de KHI carr€ (d.d.]. = 1; p <0.05), ou & Paide 2 tests de Fisher, lorsque necessaire. La comparaison des mesures vemporelles a ete faite par des tests U de Mann-Whitney bidirectionnels a p<0.05. Resultats Ramas. L.C L’analyse des enregistrements Ramas. C.S. nous a fourni des informa- Ret. spont. LC. Ret. spont. C.S tions sur l’evolution quantita- tive du ramassage des jeunes par les femelles, sur le retour spontan® des jeunes, sur les contacts mere-jeunes et sur les jeunes «restant hors du nıd», de JO1 a J19. Nombre de jeunes L’activite de ramassage Dans les conditions decrites ci-dessus, nous avons pu dis- tinguer, au cours de la periode d’elevage, trois phases (fig. 3). La premiere phase se caracte- rise par une frequence &levee de l’activit€ de ramassage. Celle-ci diminue au cours de la deuxi&me phase et disparait dans la troisieme phase. La premiere phase se prolonge jusqu’au jour J10 dans les lon- gues cages, et s’arrete au jour J05 dans les cages standard. 0 10 jours 20 Fig. 3. Nombre de jeunes ramasses ou retournant spontanement Dans les longues cRBe> au nıd, en fonction de leur äge tous les jeunes sont ramasses Structure de l’espace et retour au nid chez la gerbille de Mongolie 229 de J01 & J10. Dans les cages standard, les plus forts pourcentages de ramassage ont Et€ de 90 % le jour JO1 et de 86 % de JO2 A JO5. La deuxieme phase s’est etendue de J12 a J17 dans les longues cages et de JO6 a J14 dans les cages standard. Les frequences de ramassage sont significativement plus elevees dans les longues cages que dans les cages standard, entre J06 et J15 (les differences £tant significatives pendant chacun de ces 10 jours avec une probabilite d’au moins p<0.05). Dans les cages standard, seulement 50 % des jeunes sont ramasses par la mere ä partir du jour JO9, tandis qu’il a fallu attendre le jour J16 pour obtenir un score aussi faıble dans les longues cages. L’evolution de cette activite presente alors une diminution rapide dans les longues cages, tandis que, dans les cages standard, la diminution est reguliere et sa pente plus faible. La frequence cumulee des femelles qui n’ont pas ramasse leur trois jeunes dans les deux conditions ne se superpose A aucun moment (fig. 4). Il s’ensuit que ces conditions different significativement. Le ramassage complet s’observe sur une plus longue periode dans les longues cages que dans les cages standard. Dans les longues cages, le nombre median de jeunes non ramasses par portee a Ete nul avant le jour J16, puis de trois. Dans les cages standard, le nombre median par portee de jeunes non ramassees a et€E de un A J14, de2a ]17 etde3 a J18. Latence de ramassage Cette variable repr&sente le temps qui separe le premier contact mere-jeune du ramassage du premier jeune (fig. 5). Des analyses de varıance de Friedman faites sur ces donne&es ont montre qu’il y avait une modification dans les latences du premier ramassage quand les 10 premiers jours d’experimentation (c’est A dire quand toutes les femelles sauf une ramassent au moins un jeune) sont pris en compte (Xr = 34.9 avec d.d.1. = 9 dans les longues cages et Xr = 27.4 avec d.d.l. = 8 dans les cages standard). Afın de comparer nos resultats A ceux de WARING et PERPER, nous avons compare les latences de ramassage a JO1 & celles de JO4, puis les latences de JO4 a celles de J12. Des tests de Wilcoxon montrent qu’il y a une diminution significative entre JO1 et JO4 (Wilcoxon t = O pour les longues cages et t = 1 pour les cages standard, p <0.05) et une augmentation des latences de JO4 4 J12 (t = 7 pour les longues cages et t = 0 pour les cages standard, p <0.05). Les latences n’ont £te significativement plus elevees dans les cages standard que dans les longues cages que les jours J1O et J11. Cependant, au jour JO1l, cette latence £tait significativement plus Ale dans les longues cages que a les cages standard (Mann Whitney: U = 11, nl = 10, n2 = 9, p<0.01). Duree de ramassage C’est l’intervalle de temps necessaire pour effectuer le ramassage de trois jeunes. Cette variable fait intervenir les temps de transport, mais aussi les pauses observ£&es par la femelle, entre le premier et le dernier ramassage. Cette duree est plus longue dans les longues cages que dans les cages standard, avec une difference significative A J06 (U = 22,5,n1=n2=9, P<0.05). Le retour spontane Le retour des jeunes dans leurs nıds, par leurs propres moyens, a &t€ observ£& des le jour JO1 pour les longues cages, mais seulement au jour J11 pour les cages standard. Une forte frequence de retour spontane s’observe A partir de J11 dans les cages standard, mais seulement & partır de J16 dans les longues cages. Les frequences de retour spontane ont ete significativement plus Elevees dans les cages standard que dans les longues cages, de J08 & J18 (avec des probabilites au moins egales äp <0.05). La fig. 3 montre la proportion cumulee de portees dans lesquelles au moins un jeune a regagne le nid par ses propres moyens. saunal sap 93e,] 3p UON9UOF UI ‘SUONIPUOI xnap sap Junsey9 suep ‘o8essewei ap sauusıpmonb saousIe] sap saueIppW 'S "317 yLl c sınol OL 8 9 v 4 0 001 'S’Q sause] — 'n']89USe] Er 005 (s9PUO93S) J9us}e] "suon -IPuU09 xnap Soap Junsey> suep sounal sap Jueyuods 1N0394 191urıd np 39 oesseurei np 9fe30) uonuredsip e[ 9p 99fnuuma a9uanbauy 'F "317 0% sıno[ oL 0 S'9 'uods ']9H4 97 uods Je 'S'9 'SEWEH '9'7 'sewey sasjnuna ssausnba1g Structure de lespace et retour au nid chez la gerbille de Mongolie 21] Dans les longues cages, le nombre median de jeunes qui regagnent le nid par leurs propres moyens a Et€ nul jusqu’ä J10, et de 3 par portee A partir de ]16. Dans les cages standard, ce nombre a augmente regulierement, passant de 1 pour J08, & 2 pour J10 et 3 pour ]13. Des differences significatives dans le nombre de jeunes r&ussissant le retour au nıd ont Ete observ£es entre les deux conditions (avec p < 0.05 au moins) entre JO8 et J18. Dans les longues cages, la frequence de retour au nıd a Et d’emblee forte, tandıs que dans les cages standard, elle s’elevait progressivement. Il faut rappeler, parmi les etapes de la maturation organo-genetique, que si l’aptitude locomotrice des jeunes gerbilles est notable des J2, l’ouverture des yeux n’intervient que vers J16-]17, ce qui modifie fortement les capacites d’orientation des jeunes dans l’espace, et en particulier leur aptitude & localiser le nid. Les jeunes restant sur place Dans les longues cages, la totalite des jeunes a Et€ remasse&e par la femelle jusqu’au jour J10. Dans les cages standard, le nombre de jeunes restant sur place est passe de 10 % A J01 A 16 % a J08. Apres cette periode, le nombre de ces jeunes est demeure proche de zero, tandis que dans les longues cages, il augmentait. Cependant, aucune difference significative n’a Et€ trouvee entre les deux conditions. Discussion Des differences s’observent entre les deux conditions en ce qui concerne les frequences de comportement de ramassage, les taux de retour spontan et les taux de jeunes non ramasses. L’origine genetique commune des anımaux utilises dans cette experience nous amene & inferer les causes de ces differences A la structure de l’environnement mis ä& la disposition des anımaux. L’expression du comportement de ramassage se caract£rise, chez la gerbille de Mongo- lie, par une Evolution au cours des dıx neuf premiers jours de la vie du jeune. Cette evolution peut Etre divisee en trois phases, distinguees par les frequences de ramassage. Cecı avait et€ decrit par WARING et PERPER (1979); notre experimentation a permis de montrer qu’en plus de l’ınfluence de facteurs physiologiques, un effet modulateur des conditions environnementales peut s’ajouter. En effet, dans notre situation experimentale, les taux de ramassage des jeunes ont toujours &te plus Eleves dans les longues cages A nid bien separe que dans les cages standard de laboratoire. De plus, dans les longues cages, le comportement de ramassage s’exprime plus longtemps au cours de l’ontogenese des jeunes que dans les cages standard. Les differences de performances observees entre les deux conditions, le premier jour du test, peuvent @tre rapportees au faıt que dans les longues cages, les gerbilles adultes et leurs portees sont placees dans la cage de test la veille de la premiere manipulation, tandis que les anımaux des cages standard continuent de vivre dans leur cage familiere. Les differences entre les deux conditions se sont averees significatives essentiellement durant la periode ou le taux de ramassage £tait particulierement eleve. Cette periode a EtE plus longue dans les longues cages que dans les cages standard. Contrairement a WARING et PERPER (1979), nous n’avons pas observ& de phase ınitiale de croissance du taux de ramassage des jeunes, pendant les quatre premiers jours. Dans notre situation experimentale, l’expression du ramassage s’est revelee forte d’emblee, des le debut de la periode d’allaitement. La seule difference significative porte sur la diminution de la latence de ramassage entre JO1 et JO4-05, suivie d’une augmentation de ce parame£tre, jusqu’a la fin de notre experimentation, dans les deux conditions experimentales. 232 M. Le Berre et L. Le Guelte Comme WARING et PERPER (1979) l’avaient note, l’intensit€ des &missions ultra- sonores, quı decroit apres JO3 (DE GHETT 1974) n’est pas le seul param£tre qui explique la decroissance du taux de ramassage que ces auteurs ont observ& apres J03. Dans la situation experimentale que nous avons utilise, le ramassage s’est exprime pendant une p£riode plus longue dans les longues cages que dans les cages standard. Aucun element, bibliographique ou d’observation, ne nous permet d’envisager une &volution differentielle tant des Emissions ultra-sonores que de la thermoregulation dans chacune de ces deux conditions. Selon MacMants (1971), les jeunes gerbilles ont une regulation de type ectothermique avant J12 et deviennent progressivement endothermiques entre J12 et J20, moment du sevrage. Comme les deux categories de cages se trouvaient A la meme temperature ambiante, l’hypothese d’une chute de temperature plus importante chez les jeunes de l’une des conditions ne peut &tre retenue. Le retour spontane des jeunes au nıd s’observe plus töt dans les cages ordinaires que dans les grandes cages. Dans les premieres, la maturation musculaire permet souvent aux jeunes de rejoindre le nıd avant m&me que la femelle adulte ne puisse &tablir un premier contact avec eux. Aucun Element ne peut permettre de penser que la maturation musculaire puisse &tre ditferente dans les deux situations experimentales. De ce fait, force nous est de rapporter les differences observees aux differences inherentes aux deux conditions, et, en particulier, aux differences d’organisation de l’espace. L’activit€ de ramassage ne commence que lorsque la femelle a pergu l’absence des jeunes du nid. Quoique la distance & laquelle les experimentateurs plagaient les jeunes au moment du test etait la m&me dans les deux conditions, les femelles adultes reagissaient avec une latence plus longue au stimulus «jeune» place dans la zone «exterieure au nıd» dans les longues cages que dans les cages standard. Nos resultats peuvent £tre interpretes comme l’indice d’une meilleure perception, par les ftemelles, des limites entre ces deux zones (entre le «nid» et le «reste de la cage») dans les longues cages. Dans les cages standard, au contraire, on peut penser que l’ensemble de la cage est percu comme une extension de la zone de nıdification avec laquelle elle constitue un continuum spatial. Dans cet ordre d’ıdee, nous pensons que les longues cages constitu- ent un modele experimental qui recree des conditions de vie plus proches de celles de l’environnement naturel de la gerbille de Mongolie que les cages classiques de laboratoire. Le fait que, dans les grandes cages, l’äge auquel les jeunes arrıvent a se rendre seuls au nid est plus tardıf que dans les cages carrees s’explique par le mode de locomotion des jeunes. Dans les longues cages, les retours spontanes au nid ne deviennent vraıment frequents qu’apres l’ouverture des yeux, qui a lieu vers le seizieme jour en moyenne, dans notre elevage. L’observation montre qu’avant cet äge, le retour au nid est le resultat d’une locomotion paraissant non-orientee. La probabilite de succes de ce type de locomotion est, evidem- ment, plus forte dans une cage carr&e ordinaire puisque, dans la majorite des cas, les deplacement des jeunes se font, dans les deux conditions, en suivant, tactilement, les paroıs de la cage. Ce comportement est lıe, selon certains, a un thigmotactisme; mais ıl peut s’agır, aussi bien, du maintien d’un contact sensoriel supplementaire chez un anımal qui dispose de peu de moyens d’orientation. Les parois des cages standard ont un perimetre de 160 centimetres, les longues cages, un perimetre de 270 centimetres. Quelle que soit la direction de marche, dans les cages carr&es, le jeune aboutira au nid. Il n’en est pas de meme dans les longues cages oü une orientation en sens inverse de la direction du nid provoquera un temps de retour au nid tres long (en cas de r&orientation) ou l’absence de retour. I] est notable que, dans cette condition, le retour spontane ne depasse 50 % qu’apres l’ouverture des yeux. Ceci fait penser ä une certaine inefficacite d’un reperage tactile dans la regulation Structure de l’espace et retour au nıd chez la gerbille de Mongolie 235 de cette activite. L’olfaction ne semble pas intervenir dans la localısation du nid pendant la periode Etudiee ici (YAHR et al. 1983), et ce n’est qu’apres l’ouverture des yeux que les performances des jeunes dans les grandes cages s’ameliorent. Les performances de ramassage des jeunes et de leur retour spontane que nous avons observe&es dans les longues cages sont exactement inversement correlees. Il existe donc une concordance entre le comportement de ramassage de la femelle et l’inaptitude des jeunes ä rentrer au nid. Ceci est confirme par l’observation d’une persistance du comportement de ramassage de la femelle pendant une periode allant jusqu’a 48 jours, lorsque les jeunes d’une femelle sont remplaces periodiquement par d’autres jeunes plus immatures (Pour- TIER L., comm. pers.). L’inaptitude des jeunes pour le retour spontane dans un type d’environnement relative- ment complexe constitue un handicap vis a vis des chances de survie, dans un environne- ment naturel. Celui-ci est contre-balance par le plus fort taux de ramassage des femelles dans cette situation. Ces experiences montrent le caractere adaptatıf du comportement de la femelle, dont l’expression differentielle, en fonction des caract£ristiques de l’espace, peut se reveler d’une considerable valeur pour la survie des jeunes. Resume Les auteurs ont teste les performances de ramassage des jeunes dans deux groupes compos&s chacun de dix Gerbilles de Mongolie femelles adultes ainsi que celles de retour spontane au nid de leurs jeunes, dans deux environnements experimentaux differents, de la parturition au sevrage. Le premier environnement consistait en une tres longue cage (120 x 15 cm), munie d’une boite nichoir. Le second Etait une cage standard, carree (40 x 40 cm), avec un nıd ouvert. Les latences du premier contact femelle-jeune durant les tests ne different pas significativement. Le taux de ramassage des jeunes est plus eleve entre les jours 6 et 18 post-partum pour les anımaux des longues cages que pour ceux des cages standard. De forts taux de ramassage sont observes pendant une plus longue p£riode dans les longues cages que dans les cages standard. L’inverse est observ& en ce quı concerne le retour spontane des jeunes au nid, qui fait son apparition au dixieme jour dans les longues cages et au deuxieme jour dans les cages standard. Ces differences comportementales sont rapportees aux differences de structures des deux environnements experimentaux accessibles aux anımaux, permettant de distinguer une zone «favorable a l’elevage des jeunes» et une zone «non favorable & l’elevage des jeunes». Ce protocole a permis de montrer les oa d’adaptation du comportement de la femelle adulte & la structuration particuliere de son environnement, ce qui, dans un environnement naturel est decisif pour la survie des jeunes. Zusammenfassung Raumnutzung und Rückkehr zum Nest bei der Mongolischen Rennmaus (Meriones ungniculatus) Geprüft wurde in zwei verschiedenen Käfigtypen an je 10 Weibchen von Meriones unguiculatus, mit welcher Intensität sie ihre Jungen im Alter von 1-19 Tagen ins Nest trugen und wie weit die Jungen aktiv zum Nest zurückkehrten. Käfig A war gestreckt (120 x 15 cm) und enthielt einen geschlossenen Nestkasten. Käfig B war quadratisch und mit einem offenen Nest ausgestattet. In Käfig A wurden die Jungen intensiver eingetragen als in Käfig B. Dafür kehrten in Käfig B die jungen häufiger aktıv ins Nest zurück als in Käfig A. Diese Unterschiede können damit erklärt werden, daß Käfigtyp A deutlicher in eine Nest- und eine Außenregion gegliedert war und die Jungen bei Orientierung an der Käfıgwand mit dem Tastsınn in Käfig B eher zufällig auf das Nest stießen als in Käfig A. Bibliographie BAnnIKov, A. G. (1954): The places inhabited and natural history of Meriones unguiculatus. USSR — Acad. Sci., 410-415. BERNARD, F. (1969): Les mammiferes de Tunisie et des regions voisines — Les rongeurs. Bull. Fac. Agron. Tunis. 24-25, 39-172. BLAIR, W. F. (1941): Observations on the life history of Baiomys taylorı subator. J. Mammalogy 22, 378-383. DEGHEITT, V. J. (1974): Developmental changes in the rate of ultrasonic vocalısation in the Mongolian gerbil. Dev. Psychobiol. 7, 267-272. Duptey, D. (1974): Paternal behavior in the California mouse Peromyscus calıfornicus. Behav. Biol. 11, 247252. 234 M. Le Berre et L. Le Guelte Erwoop, R. W. (1975): Paternal and maternal behaviour in the Mongolian gerbil. Anım. Behav. 23, 766-772. HarrıeLd, D. M. (1935): A natural history study of Microtus cahfornicus. J. Mammalogy 116, 261. MARQuESs, D. M.; VALENSTEIN, E. S. (1976): Another hamster paradox — more males carry pups and fewer kill and cannıbalise youngs than do females. J. Comp. Physiol. Psychol. 90, 653-657. MacManuvs, J. J. (1971): Early postnatal growth and the development of temperature regulation in the Mongolıan gerbil. J. Mammalogy 52, 782-792. No1RoT, E. (1969): Serial order of maternal responses ın mice. Anım. Behav. 17, 547-550. WARING, A.; PERPER, T. (1979): Parental behaviour in the Mongolian gerbil-1-Retrieval. Anım. Behav. 27, 1091-1097. YAHR, P.; ANDERSON-MITCHELL, K. (1983): Attraction of Gerbil pups to maternal nest odors: duration, specificity and ovarıan control. Physiol. Behav. 31, 241-247. Adresse des auteurs: Dr. M. LE BERRE et Dr. L. LE GUELTE, Laboratoire d’Ethologie, Universite Claude Bernard, 86 rue Pasteur, F-69007 Lyon, France Z. Säugetierkunde 53 (1988) 235-244 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 The effect of fire on the small mammal community in Hluhluwe Game Reserve By A. E. Bowrann and M. R. PERRIN Department of Zoology and Entomology, University of Natal, Pietermaritzburg, Sonth Africa Receipt of Ms. 15. 1. 1987 Abstract Investigated was the effect of fire on small mammals. Four permanent trapping grids in Acacia savanna, subjected to various burning treatments, were monitored from July 1982 to December 1983. An area not burnt and an area with a patchy-burn showed an increase in rodent densities immediately after the fire. A burning experiment disclosed no anımals fleeing from the blaze. Species composition and diversity indexes of the small mammal community were relatively unaffected by the burns. No age-class of Praomys natalensis was more vulnerable to the fire than any others. Each sex of P. natalensis and Aethomys chrysophilus responded similarly to the burns while more Lemniscomys griselda females were captured ın post-burn trapping. An effect of burning was that cleanly burnt areas seemed to stimulate reproduction; relatively more P. natalensis adult came into reproductive condition during spring in these areas. The mean distance between captures (range distance) of P. natalensis fluctuated inversely with changes in population densities. The mobility of P. natalensis and L. griselda following the burns increased while the survival rates of P. natalensıs, A. chrysophilus, and L. griselda appeared greater in unburnt areas. The 1982/83 drought caused the population densities of each species in the small mammal community to “crash”. Introduction In recognition of the vital role fire plays in natural ecosystems, ıts use has now been accepted as a necessary tool in land management (EpwArps 1984). The burning pro- gramme employed at Hluhluwe Game Reserve (HGR) has two major objectives, to combat bush encroachment; and to maintain the fıre-climax grasslands (unpubl. NArTaL PARKS BOARD report). In order to maintain a rich and diverse rodent community correct grassland manage- ment in natural areas is vital. The immediate responses of small mammals to a blaze vary; anımals may flee from the flames (Deranv 1974; SwWANEPOEL 1981), or seek refuge in burrows, holes, crevices, or ıslands of vegetation missed by fire (CHEESEMAN and DELANY 1979; Bess et al. 1981) while others burn or suffocate (DELAnY 1974; CHRISTIAN 1977). CHEESEMAN and DELANY (1979) report rodents moving ahead of the flames (200-300 m), and they suggest the anımals respond to a warning stimulus (noise of combustion, smell of/or smoke) other than the heat of the fire itself. Several authors report direct mortality as being negligible because of high post-burn survival (DErany 1974; BEGG et al. 1981; SWANEPOEL 1981). Small mammal habitat is severely and abruptly modified by fire ın that food supply is reduced and cover removed. Post-burn numbers decline due to hunger and enhanced predation (BEsc etal. 1981; SwanEPOEL 1981; ROwE-RowE and LowrYy 1982). Reproduc- tion and recruitment can be affected by fire through delayed breeding and reduced litter sızes (BEGc et al. 1981). The objective of the present study was to investigate the effect of controlled burning on the abundance and diversity of small mammals in HGR. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5304-0235 $ 02.50/0 236 A. E. Bowland and M. R. Perrin Study area HGR (28° 00’ — 28° 91'S, 32° 00’ - 32° 09’E; Fig. 1) covers 23 067 ha; the topography, characterised by a profusion of drainage lines, is extremely rugged with altitude ranging from 90 m to 750 m a.s.l. The mean annual rainfall (averaged over 52 y) is 968 mm with the wettest months occurring between October and March. The study area falls within the Zululand Thornveld subcategory of the Coastal Tropical Forest Types, and the Lowveld subcategory of the Tropical Bush and .Savanna types of Acocks (1975). Forest communities are restricted from high rainfall hillsides to riverine belts. Much of the area is covered by savannas dominated by Acacias, in particular A. karroo, A. burkei, A. nılotica, A. tortılis, A. gerrardii, A. nigrescens, and A. caffra. These generally have grass cover of tall tufted perennials such as Themeda triandra and Cymbopogon spp. (BROoKs and MAcDoNALD 1983). Trapping and vegetation monitoring were conducted primarily in Acacıa savanna in the study area situated in the northeastern corner of HGR (Fig. 1). KEY: Trapping grids N 1982 burns S\ Reserve boundary DEN Study area Roads en Rivers HLUHLUWE GAME RESERVE CORRIDOR Fig. 1. A map of the 1982 burns in the Hluhluwe Game Reserve study area showing the locations of the trapping grids. 1 = clean-burn 1, 2 = patchy-burn, 3 = no-burn, 4 = clean-burn 2 The effect of fire on the small mammal community in Hluhluwe Game Reserve DO, Materials and methods Trapping was conducted on 10 x 10 grids with PVC livetraps (WırLran 1978), set at 15 m intervals. The trap was usually placed in the most likely site within one metre of the trap station; all traps were checked daily in the morning, rebaited, with a mixture of equal parts rolled oats and peanut butter, and reset if necessary. Captured animals were identified, weighed, and sexed; and capture points on the grid were recorded. The following criteria were used to evaluate reproductive condition: females, the state of the vaginal opening (perforate or imperforate) and the condition of the nipples (small, enlarged, lactating); males, the position of the testes (scrotal or abdominal). Each anımal was individually marked using a toe-clip code before release at its capture site. Animals were assigned to age-classes according to body mass ı. e. adult, sub-adult, or juvenile (Table 1). The lower limit of the adult class was calculated by subtracting the standard deviation from the mean weight of the “scrotal” or “perforate adults”, to accommodate animals in an emaciated condition caused by the drought. The mean trap distance between captures of individuals was used to assess the range distance of each species. Table 1. The weight categories used to determine age classes of four rodent species Females Males Species sub-adult juvenile sub-adult juvenile (8) (8) (8) Praomys natalensis 21-33 a <21 Saccostomus campestris 21-36 <21 <21 Aethomys chrysophilus 31-63 <31 <31 Lemniscomys griselda 21-47 al ol Population densities were estimated by a weighted mean mark-recapture formula (BEGon 1979): N ze M; xn Ca) where N = estimate of the population M; = number of marked individuals at risk on day ı m; = number of marked individuals caught on day ı n; = number of individuals caught on day ı Standard error was calculated with the formula (BEGon 1979): BE N 2m ne en Zm;+1l (£m; +1) (£m; +1)’ Species diversity was calculated using the formula (SHANNoN 1948): g-? logn - ft; logt; n where H = species diversity index, f; = number of individuals of one species caught during the trappıng session, and n = total number of individuals of all species caught during the trapping session (SHANNON 1948). Diversity indexes were compared with the test devised by HuTtcHzson (1970). After extensive exploratory trapping in the study area, four permanent grids were sited in Acacıa savanna which had not been burnt since 1980. The grids were loosely paired ı. e. about 600 m distant, the two pairs were separated by about 3.5 kms (Fig. 1). Three of the grids were burnt in the first week of August 1982. One had a patchy burn while two had clean burns; the fourth unburnt grid was used as a control. The patchy burn was hot and clean with a mosıac of discrete unburnt islands of the herb layer remaining; these islands varied in area from several to a hundred or more square metres. The clean burns were hot and thorough with no herb layer remaining. In each case the clean-burn grid was adjacent to either the patchy-burn or the no- burn grid. Grids were trapped just before (July) and just after (August) the burns in 1982 and on seven other occasions until December 1983. During the 1983 burning programme in August two grids were clean-burnt (the patchy-burn grid and its adjacent clean-burn grid). Only pre-burn trapping was conducted as low rodent numbers did not justify a post-burn trappıng programme. Vegetation recovery in clean-burn 1 grid was monitored from August to December 1982. Cover 238 A. E. Bowland and M. R. Perrin was subjectively assessed using a replicated (0,5 m?) quadrat method, which included height measured where the majority of leaves and/or inflorescences ended. In an experiment to assess the immediate response of small mammals to the blaze, a 10 m wide fire- break was burnt around a 2.7 ha grid. The following day observers were strategically placed along the fire-break before igniting the area, in order to observe and record emigrants. Results Population estimates indicate that small mammal responses to the different burning treatments vary. Post-burn density on the no-burn and patchy-burn grids increased while a reduction on both clean-burn grids was evident (Table 2). The anticipated rise in numbers in the summer breeding period was stifled by the drought; population estimates showed a Table 2. Total captures, population density estimates, diversity indexes, and trap success deter- mined from the trapping results of four grids monitored from July 1982-Dec. 1983 in Hluhluwe Game Reserve Total Population Diversity Trap Grid Month captured estimate index success (n) (animals ha’') (%) Patchy-Burn Jul Aug Sep Nov Feb Apr Aug Oct Dec Clean-Burn I Jul Aug Sep Nov Feb Apr Aug Oct Dec Jul Aug Sep Nov Feb Apr Aug Oct Dec Clean-Burn 2 Jul Aug Sep Nov Feb Apr Aug Oct Dec The effect of fire on the small mammal community in Hluhluwe Game Reserve 299 50 no - burn patchy clean 1 clean 2 ? = FIRE 35 hectare 20 per Animals Il). 5 DEFDOWENSET oO oo a3 owa3 zu «<< zw < < Fig. 2. Population estimates on the variously treated (burnt) grids from Jul 1982 to Dec 1983, standard error given above each histogram steady decline on all grids to the end of 1983, with marginal increases in the clean-burn grids in December (Fig. 2). Contingency tables indicated that there were no significant changes in species composi- tion after the burns except on the no-burn grid (Table 3). The diversity index of the small mammal community ın each burning treatment did not change significantly (p > 0.2-0.5) immediately after the fire but subsequent changes showed that the no-burn and patchy- burn grids maintained a more diverse community than the clean-burn grids (Table 2). To determine which age-class of P. natalensıs, ıf any, was susceptible to the blaze, pre- Table 3. Pre- and post-burn species composition in each burning treatment area (1982) Contingency tables (2 x x) are used to determine whether significant changes in the community as a result of fire had occurred No-burn Clean-burn 2 Patchy-burn Clean-burn 1 Species pre- post- pre- post- pre- post- pre- post- P. natalensis 47 2 33 L. griselda 2 A. chrysophilus — — S. campestris M. minutoides S. pratensıis C. hirta Total caught * significant 240 A. E. Bowland and M. R. Perrin catch increased (p < 0.05). SE distance (m) a) ® ss 2 = ° \®) distance (m) Males Differences between male and female distances are tested for significance (Student’s t) Females distance (m) itself. Table 4. Mean range distance estimates with standard errors (SE), of four rodent species in Hluhluwe Game Reserve from July 1982 to April 1983 A. chrysophilus P. natalensıis S. campestris L. griselda arid post-burn comparisons were made of each grid after establishing that the age-structure in each grid did not differ significantly from the overall age-structure. A null hypothesis that the two sets of attributes, i.e. age-struc- ture and the grid areas selected for different burning treatments, were independent of each other was upheld (x? = 2.89, p > 0.8). However, pre- and post-burn age structures of P. natalensis showed a significant difference in the clean-burn 2 treatment only, where the juvenile Pre- and post-burn sex ratios showed no significant differences for P. natalensis (x? = 0.004, df = 1, p > 0.99), A. chrysophilus (r? = 0.03, df = 1, p > 0.9) and 1. griselda (x? = 2.68, df = 1, p > 0.1). The pre- and post-burn inter- grid sex ratios of P. natalensis showed no significant differences (x? = 0.0004, df = 1, p > 0.99). The range distance for each species ıs given in Table 4, P. natalensis appears to have a margınally greater range than Saccostomus campestris, A. chrysophilus, or L. griselda. Small sample sizes precluded inter-treatment analysıs of range distances in all species but P. natalensıs. Pre- and post- burn range distance fluctuations vary inversely with changes in population density estimates (Table 5). A high percentage of pre-burn marked P. natalensıs (75.8 and 68.1 %) and Z/. griselda (66.7 %) were caught during post-burn trappıng on the clean-burn grids, while ın the no- burn grid the number of pre-burn marked P. natalensis was relatively lower at 56.8 % with only 24.1 % of the catch ın the patchy-burn grid bearing pre-burn marks. The survival rate for P. natalensis, A. chrysophilus, and L. griselda was highest in the no-burn treatment (Table 6). The ratio of breeding to non-breeding P. natalensis decreased until November (July 1:7.1, August 1:9.7, Sep- tember 1:12.5) when there was an upsurge in breeding condition and the ratio changed to 1:0.5. Analysis shows that the inter-treatment ratios vary considerably (Table 7). The recovery of vegetation cover/height is depicted in Fig. 3; despite improved habitat conditions small mammal numbers showed a temporal decline. Rainfall during the study period was well below aver- age, the effect of fire, therefore, could only be investigated in the short term, i.e. before the drought had established The experiment planned to assess the small mammal response to the blaze gave no indication of anımals fleeing from the flames; one of the 35 anımals known to be in the area fled across the firebreak. Quinn (1979) reported burnt and unburnt carcasses of small mammals found after fires in chaparral shrubs, while Tevıs (1956) actually wit- nessed animals perishing in the flames. Only one unburnt carcass (L. griselda) was found after the fire in the experi- ment during the present study. The effect of fire on the small mammal community in Hluhluwe Game Reserve 21 Table 5. Changes in pre- and post-burn population densities and mean range distances [with standard errors (SE) and a test for significance (Student’s t)] of P. natalensis in each burning treatment Treatment Population density Mean range distance estimates (anımals/ha) (m) pre-burn post-burn pre-burn SE post-burn >) 26.0 ro) ) 18.6 24.0 25) 24.8 20.9 20.8 4.3 16.5 157, 28.5 3.4 44.6 Table 6. Survival rate of P. natalensis, A. chrysophilus, and L. griselda expressed as a percentage of the number of pre-burn marked animals caught in post-burn trapping sessions Species Burning Number of marked Survival (%) treatment anımals P. natalensıis Patchy Clean 1 No-burn Clean 2 A. chrysophilus Patchy Clean 1 No-burn Clean 2 L. griselda Patchy Clean No-burn Clean 2 Table 7. The ratio of breeding to non-breeding P. natalensis in various burning treatment areas during November 1982 Treatment Breeding : Non-breeding No-burn Patchy-burn Clean-burn 1 Clean-burn 2 Discussion Factors influencing the rate at which burnt savanna grassland is recolonised by small mammals include behaviour of the anımals involved, proximity to refugia (unburnt grassland or forest inhabited during or shortly after the blaze), rate of vegetation recovery and the type of burn. Recolonisation of burnt areas may extend between two and eight or more months and can, subsequently, yield higher small mammal numbers than unburnt areas (NEAL 1970; STEwART 1972; KERN 1977; CHEESEMAN and DELANY 1979; SVANEPOEL 1981; RowE-RowE and Lowry 1982). KERN (1977), who examined the effects of annual August burns, estimated 0.62 anımals/ha immediately after the fıres, 2.16 anımals/ha two months later, and 4.17 anımals/ha in January when grasses were seeding profusely. 242 A.E. Bowland and M. R. Perrin 80 percentage cover 70 60 50 grass height (cms) 20 animals/ ha Fig. 3. Recovery of the grass sward (height and percentage cover) after an August clean-burn, related to changes in small mammal numbers, Aug 1982 to Apr 1983 The anticipated small mammal recolonisation of the burnt areas in this study did not occur despite the recovery of the herb layer, the effects of the drought are ıimplicated. Population estimates Post-burn population crashes have been recorded by many workers (NEAL 1970; KERN 1977; CHRISTIAN 1977; SWANEPOEL 1981) but pre- and post-burn trapping in the present study suggest that fluctuations in small mammal numbers following fire stem mainly from dispersal. The grids providing cover (no-burn and patchy-burn) showed an increase in numbers whereas the two clean-burn grids, where all cover had been removed, exhibited decreases. Cook (1959), however, found no evidence that surviving mice emigrated to adjacent unburnt areas, but, most studies (NEAL 1970; CHRISTIAN 1977; KERN 1981; SWANEPOEL 1981) show that migration from the burnt area to an area providing shelter is common. Species composition On the no-burn grid, where significant post-burn changes occurred, the number of P. natalensis and L. griselda were augmented by migrants from the surrounding burnt areas, while the rest of the community remained numerically constant. SwANEPOEL (1981) found that 25 % of P. natalensis migrated to an adjacent unburnt area while more mobile /. griselda vacated the burnt area completely. On the adjacent clean-burn 2 grid, P. natalensıs numbers dropped by 11.3 %. Crocidura hirta dısappeared from the patchy-burn grid, while P. natalensis numbers increased, and A. chrysophilus and L. griselda numbers remained constant. The total post- burn captures in this grid decreased yet population estimates increased; the apparent anomaly arose from the high migration rate in thıs area. First, the P. natalensis population had been considerably diluted, only 24.1 % of the post-burn captures bore pre-burn marks compared with 56.8 % in the no-burn grid. Second, increased movement was indicated by the relatively low recapture rate of allmarked anımals. It ıs clear that the small mammal community was markedly disrupted by the patchy-burn. In the adjacent clean- The effect of fire on the small mammal community in Hluhluwe Game Reserve 243 burn 1 grid the poorly represented A. chrysophilus disappeared. C. hirta numbers remained about constant and P. natalensis and L. griselda declined noticeably. Sex ratios Both sexes of P. natalensis reacted alıke to the varıous burning treatments. SWANEPOEL (1981) also reported no significant post-burn changes in sex-ratio of P. natalensis. The post-burn sex ratio of Z[. griselda, however, changed noticeably with an increase in number of females caught. The findings of SwANEPOEL (1981) too showed relatively more L. griselda females caught after the fire. Reproduction CHRrısTIan (1977) found that the effects of fire did not drastically alter the intensity or timing of breeding of deserticolous Gerbillus paeba or Desmodillus auricularis. BEcc et al. (1981) established that fire affected reproduction and recruitment in all four species in his study area which ranged from closed forest to perennial grasslands. In the present study the difference in breeding to non-breeding ratios of P. natalensis ın November 1982 on the varıous burning treatment areas suggested an influence of fire on reproduction; more adults came into reproductive condition on the clean-burn grids. The reproductive trigger may be low density but DELanY (1972) infers that the onset and termination of breeding could be correlated with biochemical and quantitative changes in diet. Range distance CHRISTIAN (1977) has reported that ıf any critical resources have been destroyed by fıre it ıs likely that survivors on the burnt area would range over greater distances in search of food and cover. The two species studied, G. paeba and D. auricularis, both had a greater range on the burnt area than on the unburnt area. Kern (1977) also found a slight increase in the home range of T. lencogaster on burnt areas. The range distances on the clean-burn treatments of P. natalensis increased after the burns while those on the no- and patchy-burn areas decreased. Though these trends are statistically not significant they do suggest that mobility is probably influenced by an interaction of the effects of fire (sub-optimal habitat, food scarcity) and population densities. Survival rate CHRISTIAN (1977) found the survival rate higher, though not significantly so, on the burnt area whereas in the present study the survival of P. natalensis, A. chrysophilus and L. griselda was higher on the no-burn area. P. natalensis showed more resilience in the clean- burn 2 area than the patchy-burn area. Perhaps the intense post-burn disruption and the onset of the drought resulted in the heavy mortality in this area. The present study showed that the small mammal community is able to cope with the fire itself as numerous aspects (sex ratios, species composition, age structure, diversity) were relatively unaffected. The incıdence of fire in the home range of a small mammal ıs a brief event and only fatal if the anımal is engulfed by the flames or asphyxiated by the fumes. Mortality, as a direct result of fire, seems to occur seldomly. The major impact of fire on the small mammal community which leads to marked declines in abundance stems from the sudden and extreme modification of the habitat whereby food supply and cover are removed. Acknowledgements We wish to thank Dr P. M. Brooks for his help in supervising this project. Natal Parks Board and University of Natal staff members willingly provided information and assistance. Numerous field 244 A. E. Bowland and M. R. Perrin assistants enabled the successful completion of many arduous tasks in the field. The CSIR, the University of Natal, and the University of Natal/Natal Parks Board Research Fund assisted financıally. Zusammenfassung Der Einfluß von Bränden auf die Kleinsäuger im Hluhluwe-Wildreservat Untersucht wurde der Einfluß von Bränden auf die Kleinsäuger. Dazu wurden vom Juli 1982-Dezem- ber 1983 die Kleinsäugerbestände von vier Probeflächen, die vollständig, teilweise oder gar nicht abgebrannt wurden, durch Markierungsfang verfolgt. Auf der nicht und der nur teilweise abgebrann- ten Fläche stieg die Anzahl der Nager unmittelbar nach dem Brand an. Bei direkter Beobachtung wurden keine vor dem Brand flüchtenden Kleinsäuger festgestellt. Die Artenzusammensetzung und Diversitätsindices blieben nach Bränden relativ unbeeinflußt. Bei Praomys natalensis waren die verschiedenen Altersgruppen nicht signifikant verschieden vom Feuer betroffen. Eine Zunahme der Weibchen, die bei Lemniscomys griselda nach dem Brand auftrat, war nicht signifikant. Auf völlig abgebrannten Flächen war der Anteil sich fortpflanzender P. natalensis erhöht. Die mittlere Aktions- raumgröße nahm bei P. natalensis mit zunehmender Dichte ab. Die Überlebensrate der Nager schien auf den nicht abgebrannten Flächen größer zu sein. Eine Dürre in den Jahren 1982/83 führte zu einer allgemeinen Abnahme aller Kleinsäuger des Beobachtungsgebietes in dieser Zeit. References Acocks, J. P. H. (1975): Veld types of South Africa. 2nd ed. Mem.Bot. Survey of S.A. No. 40. Bess, R. J.; Martın, K. C.; Price, N. F. (1981): The small mammals of Little Nourlangie Rock, Northern Territory, Australia. Aust. Wildl.Res. 8 (3), 515-528. BEecon, M. (1979): Investigating Anımal Abundance: capture-recapture for Biologists. London: Edward Arnold. Brooks, P. M.; MAacDonatp, I. A. W. (1983): The Hluhluwe-Umfolozi Reserve: an ecological case history. In: Management of large mammals in African conservation areas. Haum: Pretoria. CHEESEMAN, C. L.; DELANY, M. ]. (1979): The population dynamics of small mammal rodents in a tropical African grassland. J. Zool., Lond. 188, 451-475. CHRISTIAN, D. P. (1977): Effects of fire on small mammal populations in a desert grassland. ]. Mammalogy 58, 423427. Cook, S$. F. (1959): The effects of fire on a population of small rodents. Ecology 40 (1), 102-108. Derany, M. ]J. (1972): The ecology of small rodents in tropical Afrıca. Mamm. Rev. 2, 1-42. Derany, M. J. (1974): The Ecology of Small Mammals. Studies in Biology No. 51. London: Edward Arnold. Epwaros, P. J. (1984): The use of fire as a management tool. In: Ecological effects of fire in South Afrıcan Ecosystems. Ed. by P. de V. Booysen, N. M. Taınton. Berlin: Springer-Verlag. HurcHeson, K. (1970): A test for comparing diversities based on the Shannon formula. J. Theoret. Biol. 29, 151-154. Kern, N. G. (1977): The influence of fire on populations of small mammals of the Kruger National Park. M. Sc. Thesis, Univ. Pretoria. — (1981): The influence of fire on populations of small mammals of the Kruger National Park. Koedoe 24, 125-157. Near, B. R. (1970): The habitat distribution and activity of a rodent population in western Uganda, with particular reference to the effects of burning. Revue. Zool. Bot. Afr. 81, 29-49. Quinn, R. D. (1979): Effects of fire on small mammals in the chaparral. Calıf.-Nevada Wildl. Trans. 1979, 125-33. Rowe-RowE, D. T.; Lowry, P. B. (1982): Influence of fire on small mammal populations in the Natal Drakensberg. S. Afr. J. Wildl. Res. 12 (4), 130-139. SHANNON, C. E. (1948): A mathematical theory of communication. Bell System Techn. ]J. 27, 379-423, 623-656. STEWART, M.M. (1972): Relation of fire to habitat preference of small mammals on the Nyika Plateau, Malawı. Soc. Malawı J. 25, 33-42. SWANEPOEL, C. M. (1981): The effect of fire on a small mammal community. $. Afr. J. Zool. 16, 232-236. Tevıs, L. (1956): Effect of slash burn on forest mice. J. Wildl. Mgmt. 20, 405409. WILLAN, K. (1978): Design and field tests of a modified small mammal livetrap. S. Afr. J. Zool. 14 (2), 81-84. Authors’ address: A. E. Bowrann and M. R. Perrın, Department of Zoology and Entomology, University of Natal, P.O.Box 375, Pietermaritzburg 3200, South Africa Z. Säugetierkunde 53 (1988) 245-252 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Responses of Apennine chamois to human disturbance By I. J. PATTERsSON University of Aberdeen, Zoology Department, Newburgh, Grampian, U.K. Receipt of Ms. 13. 3. 1987 Abstract The study measured the effects of human disturbance on the behaviour of different age groups of Apennine chamois Rupicapra pyrenaica ornata in three areas with different levels of human presence in the upper Val dı Rose, Abruzzo National Park, Italy, in July 1986. There was no consistent difference in flight distance between the sexes or between grazing and resting anımals, in response to standardised experimental approaches, but yearling and sub-adult chamois had significantly shorter flight distances than had young adults. Females with kids had significantly longer flight distances than those without, although the difference was confined to resting anımals. Flight distances were least in the area with most visitors and were longest in the most remote area and there was evidence of habituation with repeated exposure to people. Introduction The population of chamois Rupicapra pyrenaica ornata ın the Italian Apennine Mountains numbers fewer than 400 animals, confined to a small area in the Abruzzo National Park, and is described as vulnerable by I.U.C.N. The animals are subject to considerable human disturbance; CEDERNA and Lovarı (1985) showed that the many visitors to the Park (2039 in one study area; 30 days’ observation) caused disruption to grazing by forcing the anımals to retreat to rock faces. Grazing was completely prevented during the mid- morning peak of tourist activity. Lovarı and RosTo (1985) found that even in the apparent absence of human disturb- ance, younger, subordinate female chamois grazed at a significantly lower rate and were significantly more vigilant than older, dominant females. Intra-group socıal rank factors were likely to be involved but it is also possible that human presence might affect the feeding of younger chamois more than that of older ones. The aims of the present study were: 1. to measure the effects of human disturbance on the behaviour of Apennine chamoıs of different age and sex, in relation to their previous behaviour (grazing or resting); 2. to compare the responses of anımals ın different areas of the Abruzzo Park with different amounts of human disturbance; and 3. to find whether chamois would habituate to the continual presence of people. Study area The main study areas were in the upper part of the Val di Rose in the Abruzzo National Park (Fig. 1), an area of limestone ridges and alpine meadows at 1850-1942 m altitude. Three areas were used: Passo Cavuto, which was visited very frequently by walkers in summer; Boccanera, which was not used by walkers but where the animals were accustomed to the frequent presence of observers; and Sterpi d’Alto, where the animals were approached less frequently. A few observations were also made on Mt. Amaro, 4km to the NW, where the animals were less accustomed to people. The study was carried out during July, 1986. Each of the main study areas supported a largely separate flock of up to 30 chamois, some of which had been ear-tagged for individual recognition in earlier studies. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5304-0245 $ 02.50/0 246 1. J. Patterson j Om Mt. Sterpi d’ Alto Mt. Boccanera Fig. 1. The main study areas (shaded) in the upper Val dı Rose, Abruzzo National Park, L’Aquila, Italy Material and methods Animals were allocated to age classes, using the length of their horns in relation to ear length (Lovarı 1985): yearlings had horns around or a little less than the length of their ears; sub-adults (2-3 years old) had horns Y4-Y3 longer than the ears; young adults (4-5 years old) had horns 1% times ear length while adults (over 5 years of age) had horns at least twice the length of the ears. In animals at least two years old, males could be distinguished by their thicker, more strongly curved horns and their penile hair tuft. Yearling males and females were not distinguished. Some naturally-occurring disturbance of chamois by visitors was observed at close range but such incidents were highly variable in the number of people involved, their direction and speed of approach, whether they were noisy or quiet, etc., so the main study used standardised approaches by the observer. After an initial acclimatısation period of at least 10 min an anımal wıth no others between it and the observer was selected and its sex, age class and activity (grazing or resting) were noted. It was then approached across the slope at a slow walk (0.25 m/s), avoiding any noise or sudden movements. The distance between the animal and the observer was measured with a range-finder whenever there was a change in the chamois’ behaviour and the approach was suspended immediately the animal began to move away. The observer then retreated before starting to approach a new anımal. As far as possible, only one approach was made to each anımal in a flock on each day, but anımals without ear tags were undoubtedly approached on different days, leading to some non-independence in the data and consequent need for caution in the interpretation of statistical tests. Where repeat approaches were carried out on ear-tagged anımals, a mean value for each has been used. Results Behavioural responses to human approach In almost all of the 225 approaches made in the main study areas the anımals showed the same sequence of behaviour; grazing animals stopped feeding, oriented their heads towards the observer (noted as the alert distance) and moved away, usually with their tails raised (noted as the flight distance); similarly, resting animals oriented and rose to their feet before moving off. Only 3 % omitted orientation of the head before moving. Many (45 %) of the 31 resting animals which were ruminating steadily before being approached continued to do so until they moved off and a further 29 % even continued to ruminate as they moved. Alarm snorts were given ın only 4.5 % of approaches. Responses of Apennine chamoıs to human disturbance 247 The changes in the anımals’ behaviour during approaches tended all to occur within a few seconds, at the same distance from the observer; in only 23 % of approaches did the anımal stop activity and orient to the observer at a distance greater than the eventual flight distance. The proportion doing so was, however, significantly higher in resting than in grazing anımals and was slightly but not significantly greater in females accompanied by kids than in other adults females (Table 1). In such animals the mean difference between Table 1. Percentage of animals with alert distance greater than flight distance Grazing Resting Alert % Alert Without Kids Pass 18 Boccanera | 5 Sterpi d’Alto 3 Total : 26 %=27.42; p< 0.001 With Kids Pass Boccanera Sterpi d’Alto Total 25.0 =2.10; NS Overall, 22.7 % with alert distance greater than flight distance. the alert and flight distances was 1.53 + 0.15 m (grazing), 2.02 # 0.25 m (resting) and 3.11 + 0.56 m in females with kids. (There were no statistically significant differences, however, between these distances.) Since the majorıty of anımals became alert and then moved away without further approach by the observer, flight distances alone were used ın most of the following analyses. Flight distances in relation to previous activity This could be compared in nine categories of anımal (excluding females with kids); in adult females on Sterpi d’Alto the flight distance was significantly higher ın grazing than in resting anımals but there was no significant or consistent difference in any of the other groups (Table 2). Marked animals approached both while grazing and while resting also showed no consistent difference in flight distance. Alarm snorts, although uncommon, occurred in 10.2 % of 59 approaches to resting anımals but in only 2.1 % of 143 to grazing ones (X° = 6.39; p <0.01). They occurred in a quarter of the eight tests on Mt. Amaro. Among females accompanıed by kids, resting anımals had significantly higher flight distances than had grazing ones ın the Pass and Boccanera areas (Fig. 2). On Sterpi d’Alto, grazing females with kids had flight distances as long as those of resting ones. Flight distance in relation to the sex of the animal This could be compared in only fıve categories of sub-adult, since the sex of yearlings was not determined and adult males were uncommon in the study area. There was no consistent tendency for one sex to have a greater flight distance (Table 3). Subsequent sections will therefore combine grazing and resting anımals of both sexes (excluding females accompanied by kids). 248 I. J. Patterson Table 2. Flight distance in grazing and resting animals Age/Sex Grazing Resting Z Yearlings : Pass Rass Sterpi Pass : Boccanera Sterpi Pass Pass Sterpi mo DOWN UND *t=2.86; p=0.017; no other significant differences. Mean flight distance (m) No With kids kids kids kids i kids Graze Rest Graze Rest Graze Rest Passo Boccanera Sterpi d’ Alto Fıg. 2. Flight distances in adult females with and without kids. Asterisks above the columns indicate significant differences from females without kids (*- p <0.05; **"*"- p <0.001, t tests). Asterisks within columns indicate significant differences between grazing and resting animals (p <0.05, t tests) Table 3. Flight distance in male and female sub-adults Activity Female Pass grazing Pass resting Boccanera grazing Sterpi d’Alto grazing Sterpi d’Alto resting No significant differences. Responses of Apennine chamoıis to human disturbance 249 Flight distance in different age classes Contrary to expectation, young chamois were not more sensitive to disturbance than were older ones - yearlings and sub-adults had slightly but significantly shorter flight distances than had young adults (Fig. 3). Analysis of variance, however, showed no significant variation in flight distance over the four age groups taken together. = 15 ® o ie < (ap) 7) = 10 te = vun e 83 . . . . . ® Fig. 3. Flight distance in relation 0 S age. Grazing and resting anımals of both sexes have been combined, omitting only those with kids. Young adults vs. sub-adults and yearlings, p gen 34 49 23 25 <0.05 (t tests) Yearling Sub-adult Young-adult Adult Age class Resting females with kids had longer flight distances than females without kids in all of the three main study areas (Fig. 2). Three ear-tagged females, approached while resting away from their kids, all had shorter flight distances than when resting with their kids. Flight distance in different areas In all age groups there was significant varıation in mean flight distance between the three main study areas (ANOVA; p <0.01), with the shortest distances ın the heavily visited Pass area and the longest in Sterpi d’Alto, where the anımals were visited least (Fig. 4). The moderately-studied Boccanera area was intermediate, but young adults there had signifi- cantly longer flight distances than those in the Pass area. One marked adult female seen in two areas had a longer flıght distance when she was on Sterpi d’Alto than when she was ın the Pass. The small number of observations made on Mt. Amaro suggested that flight distances there were about twice those at the Pass (Fig. 4). Flight distance with repeated approach When the approaches made to a given sex and age category in each study area were divided into their earlier and later halves, all seven categories wıth sutficient data showed a shorter flight distance in the second half of the study compared to the first, with significant differences found among yearlıngs and subadult males (Table 4). Overall, the mean reduction in flıght distance was 1.7 m. Sıx marked anımals approached more than once over the study period showed no consistent change ın flight distance, but successive approaches were usually at intervals of several days. However, in four anımals approached two or three times in quick succession, only two showed a reduction in flıght distance while one stayed the same and one showed an increase. 250 I. J. Patterson Yearlings Discussion Yearlıng and sub-adult chamois, although found by Lovarı and Rosto (1985) to be more vigilant than older ones, did not be- come alert and flee from a quietly-approach- ing person at greater distances than did older animals; indeed they allowed significantly closer approach than did young adults. This supports Lovarı and Rosrto’s (1985) sugges- tion that the vigilance of younger anımals may be mainly social, with attention directed at other chamois. It is also possible that the anımals were looking out for other potential predators such as canıds and were not con- cerned about people. There may also be a higher level of curiosity in younger animals, counteracting fear and leading to their stay- ing longer while being approached. Resting anımals might be expected to have shorter flight distances than grazing ones, which can move away easily, while the rest- ing ones have first to rise to their feet, which might require a higher threshold of fear to be exceeded. However, most chamoıs showed no difference in flight distance with previous activity, wıth only adult females on Sterpi d’Alto having significantly shorter flight dis- tances while resting. The opposite was true for females accompanied by kids in the Pass and Boccanera areas. In these, the longer Fig. 4. Flight distance in different study areas; flight distances of resting animals may be categories of anımal as in Fig. 3. Single asterisk related to a greater vulnerability of sitting indicates significant difference from Pass 5 re Id (p<0.05; t tests); double asterisk indicates sig- animals to a sudden rush by predator. This, nıfıcant difference from Pass and Boccanera however, explains only the animals’ rising to (p<0.01, t tests) their feet as a “precautionary” measure, not Mean flight distance (m) Passo Bocca Sterpi Amaro * D=0:052 Table 4. Change in flight distance with repeated approach Activity Area First half of approaches Second half ofapproaches t N SE Pass 10 Pass 8 Pass 10 Pass 8 Pass 8 Pass 9 Sterpi 7 1 DDDFODFD 2 9 ge 5 5 Difference in the same direction in all 7 sets; p< 0.05, Sign Test. Responses of Apennine chamoıs to human disturbance 251 why they go on to move away earlier than grazing anımals, unless having risen somehow predisposes them to move. The longer flight distances ın resting females with kids, compared to those without, may reflect the vulnerability of the young, which were only a month old and had just emerged on the meadows from the clıff nursery areas. CEDERNA and LovaRrt1 (unpublished) found sımilarly that alert distance was significantly greater in flocks containing a large proportion of kids. Flight distances decreased in the course of the study, presumably as the anımals habituated to the same person moving quietly among them in a predictable way. It is also possible that the observer was learning how to approach more effectively, but care was taken to maintain a uniform technique throughout the study. The failure of ear-tagged and other individuals approached several times to show any consistent decrease in flight distance may be due to the small number of approaches to each anımal and the long gaps between some of the successive tests. MCLAREN and GREEN (1985) similarly found no consistent effect of repeated approaches to musk oxen Ovibos moschatus. Habituation effects can explain the differences in flight distance between areas, with the shortest in the area most visited by people (Pass) and the longest in the most remote area (Mt. Amaro), where there was also the possibility of some poaching ($S. LovaRı, pers. comm.). In the main study areas, CEDERNA and Lovarı (unpublished) showed a decrease in mean flight distance from 25 m ın 1977-78 to 19 m in 1981-82. This reduction has apparently continued, to the 11 m found in the present study, presumably as the anımals have continued to habituate to the close proximity of people. The results of this study are encouraging for the conservation of chamois; younger animals appear not to be more affected by disturbance than older ones, as had been feared, and flocks ın close proximity to heavily visited areas appear to be habıtuating progressively to human presence. Recent improvements in visitor control in the Val di Rose, whereby ın the busiest period (July and August) visitors are mainly confined to guided parties restricted to the marked trails by increased wardening, seem greatly to have reduced the kind of harassment of the anımals described by CEDERNA and Lovarı (1985). Continued “benign” exposure to people should encourage further habituation of the anımals and so reduce the effects of visitors on them. Acknowledgements I am indebted to Professor SANDRO LovaRrt for his invaluable help and advice and to the Director of the Abruzzo National Park for permission to work there. The study was supported financıally by Aberdeen University, The Carnegie Trust and The Royal Society. Zusammenfassung Reaktionen apenninischer Gemsen auf menschliche Störung Diese Untersuchung befaßte sich mit der Wirkung menschlicher Störung auf das Verhalten verschie- dener Altersgruppen von apenninischen Gemsen, Rupicapra pyrenaica ornata, in drei Regionen mit verschieden häufiger menschlicher Anwesenheit im oberen Val dı Rose, Abruzzo Nationalpark, Italien, im Juli 1986. Bei standardisierten Annäherungsversuchen wurde kein signifikanter Unterschied bei den Flucht- distanzen zwischen Männchen und Weibchen oder zwischen grasenden und ruhenden Tieren gefun- den, aber die Fluchtdistanzen von einjährigen und sub-adulten Gemsen waren statistisch bedeutend kürzer als die von jungen Adulten. Weibchen mit Jungen hatten statistisch größere Fluchtdistanzen als Weibchen ohne Junge, obwohl dieser Unterschied nur ruhende Tiere betraf. Die Fluchtdistanzen waren am kürzesten in der am häufigsten besuchten Region und am größten in der abgelegensten. Es gab Anzeichen dafür, daß wiederholte Begegnung mit Menschen zur Gewöhnung führte. DD I. J. Patterson References CEDERNA, A.; LovaRrı, $. (1985): The impact of tourism on chamoıis feeding activities in an area of the Abruzzo National Park, Italy. In: The Biology and Management of Mountain Ungulates. Ed. by S. Lovarı. London: Croom Helm. 216-225. Lovarı, $. (1985): Behavioural repertoire of the Abruzzo chamois, Rupicapra pyrenaica ornata Neumann, 1899 (Artiodactyla: Bovidae). Säugetierkundl. Mitt. 32, 113-136. Lovarr, $.; RosTo, G. (1985): Feeding rate and social stress of female chamois foraging in groups. In: The Biology and Management of Mountain Ungulates. Ed. by $S. Lovarı. London: Croom Helm. 102-105. MCLAREnN, M.; GREEN, J. (1985): The reactions of musk-oxen to snowmobile harassment. Arctic 38, 188-193. Author’s address: Dr. Ian J. PATTERsON, Zoology Department, Culterty Field Station, Aberdeen University, Newburgh, Grampian, U. K., AB4 OAA Z. Säugetierkunde 53 (1988) 253-255 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 NZSSEINSIE EIENEITELSENEFERSUR ZNIITTRETE UNE First record of the Pygmy killer whale, Feresa attenuata Gray, 1875 from Peru, with asummary of distribution in the eastern Pacific By K. Van WAEREBEER and J. C. REvEs Laboratorinm voor Morfologie en Systematiek, Museum voor Dierkunde, Rijksuniversiteit Gent, Belgium, and Grupo Cetaceos, ECCO, Asociaciön de Ecologia y Conservaciön, Lima, Peru Receipt of Ms. 10. 11. 1987 The distribution of the pygmy killer whale, Feresa attennata Gray, 1875 was reviewed by PERRIN and Husss (1969) and by Ross and LEATHERWOOD (in press), and is assumed to be circumglobal in tropical and subtropical waters. The first record of this species from the eastern Pacific was a juvenile captured in May 1967 during commercial tuna fishing operations 300 to 400 nautical miles off Costa Rica (PERRIN and Huzgs 1969). National Marine Fisheries (NMFS) observers aboard U.S.-registered tuna purse seiners operating in the eastern tropical Pacific reported 25 sıghtings of pygmy killer whales in the period 1971-1985 (as summarized in the Figure; PERRIN, pers. comm.). Feresa ıs not seen ın the northern part of the tuna grounds, towards the Gulf of California. Only three sıghtings are from south of the equator, where NMFS survey effort has been comparatively low: 25 anımals at 02° 32’S, 94° 11’W on 17 January 1979; 9 at 07° 20’ S, 85° 13’ W on 6 December 1980, and 15 at 02° 27’ S, 109° 58’ W on 15 March 1981. In addition, during the IWC/ IDCR research cruise in the eastern tropical Pacific in November and December 1982, one school of 8 animals without calves was seen at 08° 37’ S, 88° 04’ W (Donovan 1984). Data are presented below on the first report of the pygmy killer whale from Peru, the most southerly record of this species in the eastern South Pacific. On 30 November 1984 the mummified remains of a pygmy killer whale were disco- vered ın one of the several dumps in the desert surrounding Pucusana, a small fishing town in central Peru (12° 30’ S, 76° 48’ W). At the same place, many tens of skeletons of several other small cetaceans were found, all victims of the Peruvian small cetacean fishery. In Pucusana most small cetaceans are caught ın gill nets, wıth the remainder captured by a varıety of other methods (Reap et al. 1985; VAn WAEREBEEK and REYEs 1986; Van WAEREBEER et al. 1987). The majority of local fishing occurs well within 100 nautical miles of shore and mostly much closer. The Feresa attennata specimen we report consists of the head, the hyoids, sıx posterior lumbar vertebrae, the complete series of caudals (32), 23 chevron bones, both pelvic bones, the flippers, and the dorsal fin. The intact skull, covered by mummified skin, showed the rounded head typical of thıs species; all underlying soft tissue had disappeared. Sex and total length of the anımal could not be determined. Inıtially the anımal was thought to be physically immature considering the small size of the skull (339 mm condylobasal length) compared to the 352-405 mm range (X = 373 mm, n = 27) listed by Ross and LEATHERWOOD (in press). However, fusion of the epiphyses of the caudal and lumbar vertebrae as well as in the flıpper bones, the near closure of tooth pulp cavities, the flattening of the dorsal surface of the rostral portion of the premaxillae lying on the same U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5304-0253 $ 02.50/0 254 K. Van Waerebeek and ]J. C. Reyes 160° 140° 120° 100° 80° Summary of known distribution of Feresa attenuata in the eastern Pacific: @ = sightings by NMFS observers between 1971-1985 (PERRIN, pers. comm.); + = sighting during IWC/IDCR research cruise (Donovan 1984); 1 = specimen collected off Costa Rica (PERRIN and Husss 1969); 2 = specimen from Pucusana, Peru (present paper). Central Pacific records (e.g. Hawaii) are not shown here level as the maxillae (sensu FRAsER 1960), and the general degree of fusion of the cranial sutures all indicate that the whale was at least approaching physical maturity. Cranıal measurements (in mm) according SCHNELL et al. (1982), and as a percentage of the condylobasal length (CBL) shown in parentheses, are: CBL, 339 (100); length of rostrum from base, 161 (47.5); length of rostrum from pterygoid, 203 (59.9); wıdth of rostrum at base, 122 (36.0); width of rostrum at Y length, 98 (28.9); width of rostrum at % length, 86 (25.4); width of premaxillaries at '% length, 57 (16.8); wıdth of rostrum at % length, 69 (20.4); preorbital width, 203 (59.9); postorbital width, 233 (65.8); skull width at zygomatic process, 224 (66.1); skull width at parietals, 164 (48.4); height of braincase from basısphenoid to summit of supraoccipital, excluding crest, 113 (33.3); internal length of braincase, including occipital condyles, 138 (40.7); maximum width of premaxillarıes, 85 (25.1); greatest width external nares, 46 (13.6); length mesethmoid, 60 (17.7); greatest width of left temporal fossa, 63 (18.6); orbital length, 59 (17.4); length of left antorbital process, 44 (13,0); maximum separation of pterygoids, 1 (0.3); greatest wıdth of internal nares, 58 (17.1); length of left tympanic cavity, 64 (18.9); length of right tympanic cavity, 63 (18.6); greatest distance between left and right pterygobasioccipital sutures, 59 (17.4); length left upper tooth row, 108 (31.9); number of teeth-upper left, 10; -upper right, 9; -Jower left, 11; -lower right, 12; length of left lower tooth row, 120 (35.4); greatest height of left ramus, 72 (21.2); tooth width, 5.8; greatest length of left ramus, 269 (79.3). Both left and right pterygoid sinuses were eroded in “basket like” lesions as associated with infections of the nematode, Crassicauda sp. The damage to the bone could be diagnosed as severe (PERRIN and Powers 1980). The specimen is provisionally kept by the authors at their cetacean study collection ın Lima under no. KVW-032. The discovery of a pygmy killer whale combined with several other recent findings of pelagic, warm-water adapted small cetaceans on the coasts of Peru and Chile (Van First record: of the Pygmy killer whale from Peru 255 WAEREBEEK and REyEs 1986; GUERRA et al. in press), suggest that the Humboldt Current is more a heterogenous and highly variable water body than a continuous and impenetrable cold water barrier. Acknowledgements W.F. PERRIN, G. J. B. Ross and $. LEATHERWOOD are thanked for generously supplying unpublished data. The same persons and P. J. H. van BREE made several very useful comments on a previous draft of this note. Funding came from a 1984 research grant to K. Van WAEREBEER by the “Leopold III- Fonds voor Natuuronderzoek en Natuurbehoud” (Brussels), through A. CAparT and J.-P. Gosse of the “Institut Royal de Sciences Naturelles de Belgique”. Logistical support was provided by B. A. LuscomBE of ECCO and by J. G. Mean and C. W. PoTTER of the Smithsonian Institution, Washington, D.C. Literature Donovan, G. P. (1984): Small cetaceans seen during the IWC/IDCR research cruise in the eastern tropical Pacific, and in particular off Peru. Rep. Int. Whaling Comm. 34, 561-567. FRASER, F. C. (1960): A specimen of the genus Feresa from Senegal. Bull. Inst. fr. Afr. noire (A) 22, 699-707. GUERRA, C.; WAEREBEEK, K. Van.; PORTFLITT, G.; Luna, G. (in press): The short-finned pilot whale, Globicephala macrorhynchus Gray, 1846, a new record for Chile. Estudios Oceanolögicos, Universidad de Antofagasta, Chile. PERRINn, W. F.; Husgs, C. L. (1969): Observations on a young pygmy killer whale (Feresa attenuata Gray) from the eastern tropical Pacific Ocean. Trans. San Diego Soc. Nat. Hist. 15, 297-308. PERRIn, W. F.; Powers, J. E. (1980): Role of a nematode in natural mortality of spotted dolphins. ]. Wildl. Manag. 44, 960-963. READ, A. ].; WAEREBEEK, K. Van,; REYES, J. C.; McKınnon, ]. S.; GAsKIn, D. E. (1985): Preliminary observations on the catch of small cetaceans in Peruvian coastal waters. Abstracts, Soc. Mar. Mammalogy, Sıxth Bien. Conf. Biol. Mar. Mammals, Vancouver, 22-26 November 1986. Ross, G. J. B.; LEATHERWOOD, S$. (in press): Pygmy killer whale Feresa attenuata Gray, 1874. In: Handbook of Marine Mammals. Vol. 4. Ed. by S. H. Rıncway; R. Harrıson. New York, London: Academic Press. SCHNELL, G. D.; Doucras, M. E.; HoucH, D. J. (1982): Geographic varıation in morphology of spotted and spinner dolphins (Stenella attennata and S. longirostris) from the eastern tropical Pacific. NOAA, Nat. Mar. Fish. Serv. L]-82-15C, 1-213. WAEREBEEK, K. Van; Reyes, J. C. (1986): The first records of the short-finned pilot whale Globicephala macrorhynchus for Peruvian waters. Abstracts, 14th Symp. European Assoc. Aquat. Mammals, Barcelona, 16-19 March 1986. WAEREBEEK, K. Van; REyEs, J. C.; LuscoMBE, B. A. (1987): A second record of the Pygmy sperm whale Kogia breviceps (de Blainville, 1838) (Cetacea, Physeteridae) from the Peruvian coast. Z. Säugetierkunde 52, 192-194. Authors’ address: KoEN VAN WAEREBEEK and ]J. C. REyEs, Grupo Cetäceos-ECCO, Vanderghen 560, 2A, Lima-27, Peru BUIEHBESBRITEE ERWIN GRZIMER, B. (Hrsg.): Grzimeks Enzyklopädie Säugetiere - Band 4. München: Kindler- Verlag 1987. 648 S., zahlreiche Abb., Lexikon-Großformat. Leinenausgabe: Subskrip- tionspreis DM 128,-, später DM 148,-, ISBN 3-463-42004-X; Luxusausgabe (Halbleder): Subskriptionspreis DM 168,-, später DM 198,- ISBN 3-463-42104-6 Mit „Grzimeks Enzyklopädie Säugetiere“ ist ein voluminöses, fünfbändiges Werk konzipiert, das in Erweiterung der vier 1970 erschienen Säugetierbände aus „Grzimeks Tierleben“ andere Akzente setzt. Neben einer verständlichen Vermittlung allgemeiner biologischer Sachverhalte über Säugetierarten sind vor allem neuere Erkenntnisse aus den Gebieten Stammesgeschichte, Okologie und Ethologie Anliegen der Abhandlungen sowie Häufigkeit und Bedrohung der Arten und ihre Beziehungen zum Menschen. Aus redaktionellen Gründen liegt als erstes nun der 4. Band vor. Er behandelt in Fortsetzung der Carnıvora die Pantherkatzen und Verwandte, die Canıdae und die Pinnipedia, des weiteren die Ordnungen Lagomorpha, Cetacea, Tubulidentata, Proboscidea, Hyracoidea, Sirenia, Perissodactyla. Die Beschreibungen der Ordnungen erfolgen nach einheitlicher Gliederung in Einlei- tung, Stammesgeschichte und Abhandlung über die allgemeine oder besondere Biologie der Arten. Den Kapiteln vorangestellt sind sog. Basisinformationen, die kurzgefafßt einen Überblick über gruppenspezifische Besonderheiten darlegen. Zusätzlich ermöglichen tabellarisch gegenübergestellte Angaben über Körpermaße, Merkmale, Fortpflanzung, Lebenslauf, Nahrung, Feinde, Lebensweise und Lebensraum schnelle und detaillierte Information über einzelne Arten. Im Text wird auf ältere und neuere Erkenntnisse biologischer Besonderheiten eingegangen, ferner auf die Bedeutung vieler Arten in der Mythologie der Völker, ihre Nutzung durch den Menschen in Vergangenheit und Gegenwart, rezente Bestandsentwicklungen, Maßnahmen bei Bestandsbedrohungen und vieles andere mehr. Verbreitungsskizzen der meisten Arten und eine besonders attraktive und beeindruckende Ausstattung mit Sıruationsaufnahmen aus freier Wildbahn unterstützen die Beschreibungen. Fremd- wörter und Fachausdrücke wurden vermieden oder verständlich, wenn auch nicht immer einheitlich erklärt. Die Bearbeitung der Arten und systematischen Gruppen dieses Bandes ist auf 23 Autoren aufgeteilt. Die Abschnitte fallen dementsprechend unterschiedlich aus sowohl in der Form der Darstellung als auch in der inhaltlichen Gewichtung. Während in einigen Kapiteln eine stärker erzählende, mit eigenen Erlebnissen angereicherte Darstellungsform überwiegt, wird ın anderen auf knapp gefaßßte und sachliche Vermittlung von Information Wert gelegt. Die Mehrzahl der taxonomi- schen Gruppen wurde von wenigen Autoren bearbeitet, so daß diese Unterschiede kaum als störend empfunden werden. Die Kapitel wirken in sich geschlossen und machen einen abgerundeten Gesamt- eindruck. In der Regel sind alle bekannten Arten taxonomischer Gruppen erwähnt und unserem Wissensstand entsprechend mehr oder weniger ausführlich gekennzeichnet. Bei den Lagomorpha stehen allerdings die Arten Zepus europaeus und Oryctolagus cuniculus übergewichtig im Zentrum. Wenig Information erhält man in dieser Ordnung über andere Arten der Leporidae, und auch die Ochotonidae sind vergleichsweise kurz dargestellt. Auf besonders breitem Raum von über 100 $S. und mit einer besonders starken „Zersplitterung“ der Bearbeitung unter mehreren Autoren, die in abwechselnder Reihenfolge zu Wort kommen, sind die Canıdae abgehandelt. In diesem Abschnitt können an mehreren Stellen Wiederholungen, inhaltliche Widersprüche und als wissenschaftlich belegt ausgewiesene, spekulative Ansichten zu Verunsicherungen führen oder als störend empfunden werden. Ferner zeigt der Band in bestimmter anderer Hinsicht eine etwas unausgewogene Konzeption oder Bearbeitung, denn ein Leser, der an Haus- und Nutztieren interessiert ist, wird vergleichsweise detailliert über Haushunde informiert, in deutlich geringerem Maße über Hausesel und Hauspferde, so gut wie überhaupt nicht jedoch über Hauskaninchen. Unerwähnt bleibt auch die Bildung zahlreicher Farbschläge von Farmfüchsen. Auf die Nutzung von Rot- und Eisfuchs als Pelztiere wird nur kurz in unbewiesener und emotionaler Weise, der aktuellen Diskussion um Pelztierhaltung kaum förderlich, eingegangen. Im Hinblick auf den Umfang des Bandes fallen die angeführten Mängel jedoch kaum ins Gewicht, denn insgesamt liegt eine schöne und informative Dokumentation auf modernem Wissensstand vor, die ein Verständnis für die besondere Biologie der behandelten Arten fördert. Das Anliegen des Verlages, ein Tierlesebuch, ein Tierbilderalbum und ein Tierlexikon in einem Werk zu erstellen, kann grundsätzlich als gelungen bewertet werden. Es wird sicherlich von interessierten Laien und Fachleu- ten mit Gewinn genutzt werden. D. Kruska, Kiel NATUR + RECHT Zeitschrift für das gesamte Recht zum Schutze der natürlichen Lebensgrundlagen und der Umwelt NATUR + RECHT ist im deutschen Sprachraum die erste spezielle Zeitschrift für das gesamte Recht zum Schutze der natürlichen Lebensgrundlagen. Sie willdem Naturschutz- und Landschaftspflegerecht in Wissenschaft, Ausbildung und Praxis zu der Bedeutung verhelfen, die ihm als Kernstück des Umweltschutzrechts zukommt. NATUR + RECHT macht deshalb die Gemeinsamkeiten landesrechtlicher Sonderregelungen transparent und faßt die Publikationen auf verwandten Rechtsgebieten zusammen, die sich auf eine Vielzahl von Rechts- und interdisziplinären Zeitschriften mit häufig nur regionaler Bedeutung verteilen. NATUR + RECHT setzt als Rechtszeitschrift die sachliche Diskussion in Gang und belebt sie, verbessert Kommunikation und Argumentation und trägt somit zur Rechtsfortbildung bei wie auch zum Aufbau des oft beklagten Vollzugsdefizits. « NATUR + RECHT. Zeitschrift für das gesamte Recht zum Schutze der natürlichen Lebensgrundlagen und der Umwelt. Herausgegeben von Claus Carlsen, Dieter Engelhardt und Ludger Pielow. Erscheinungsweise: 8 Hefte pro Jahr. Abonnementspreis (1988): jährlich 248,— DM zzgl. 10,50 DM Versandkosten PAJL (Inland), bzw. 12,— DM (Ausland). Einzelheft 34,— DM P/REV Verlag Paul Parey : Hamburg und Berlin Erscheinungsweise und Bezugspreis 1988: 6 Hefte bilden einen Band. Jahresabonnement Inland: 308,- DM zuzüglich 13,80 DM Versandkosten, Gesamtpreis 321,80 DM einschließlich 7% Mehrwert- steuer. Jahresabonnement Ausland: 308,-— DM zuzüglich 18,-— DM Versandkosten. Das Abonne- ment wird zum Jahresanfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, Spitalerstraße 12, D-2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesellschaft für Säugetier- kunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. Z. Säugetierkunde 53 (1988) 4, 193-256 GRZIMEKS ENZYKLOPÄDIE 20 Jahre nach „Grzimeks Tierleben“ völlig neu: GRZIMEKS ENZYKLOPÄDIE - SÄUGETIERE in 5 Bänden 3240 Seiten im Lexikonformat bieten dem Zoologen - jüngste Erkenntnisse der Feldforschung, dargestellt von Autoren, die diese Befunde oft selbst gewonnen haben, wie Jane Goodall, George B.Schaller, Dian Fossey, John E Eisenberg, Valerius Geist, Kosei Izawa u.v.m.; - neueste Daten zu Anatomie, Physiologie, Ethologie und Ökologie der Säuger-Spezies in vergleichend-tabellarischer Form; - 450 Verbreitungs- bzw. Ausbreitungskarten; - 3000 Farbfotos und 500 Zeichnungen. - In einem Anhang: Gesamtregister, Tierwörterbuch, zoologische Systematik - ein »Muß« für jeden Zoologie-Studenten! So urteilen Wissenschaft und Presse: »Für zoologische Systematiker und Evolutionsforscher, die Säugetiere in ihre Arbeit einbeziehen, wird das Werk in den nächsten Jahrzehnten ein Quell wichtiger Informationen sein« (Prof. Wolf Herre, Kiel). »Die Farbfotos - ein Augenschmaus« (»Natur«) »Führende Zoologen aus aller Welt brachten ihre aktuellen Erkenntnisse zu Ökologie, Verhalten und Evolution in dieses Werk ein, und man „beißt“ sich beim Durchblättern der Bücher unweigerlich fest, weil die Texte A Don an geschrieben sind und man die jahre- lange Felderfahrung der Autoren spürt.« Entomofauna, Zeitschrift für Entomologie Subskriptionspreis der Leinenausgabe pro Band DM 128,- Späterer Preis DM 148,-; Subskriptionsersparnis je Band DM 20,-; Darum: jetzt bestellen! (Es gibt außerdem eine Halblederausgabe.) Ihr Buchhändler zeigt Ihnen gern Band IV im Original. Achtung: Die Subskription endet am 30.11.1988! KINDLER VERLAG MÜNCHEN | Vol. 53 (5), 257-320, Oktober 1988 ISSN 0044-3468 C21274F7 ZEITSCHRIFT FÜR SAUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Iraan der Deutschen Gesellschaft sur SOlSBo N RL 700 De N Zud/ | \ 7 oa 200 } Mamm \ / Nolz, Irmhild: Ergebnisse automatischer Aktivitätsaufzeichnungen. an/ffächenstubenkolonien de Bechsteinflede-- maus (Myotis bechsteini). — Results of automatically monitoring "Bächstein.bats-aetivities 257 aisbert, J.; Löpez-Fuster, M. J.; Garcia-Perea, Rosa; Ventura, J.: Distribution and biometry of Sorex granarius (Miller, 1910) (Soricinae: Insectivora). — Verbreitung und Biometrie von Sorex granarius (Miller, 1910) (Soricinae: Insectivora) 267 3lood, B. R.; McFarlane, D. A.: Notes on some bats from northern Thailand, witn comments on the subgeneric status of Myotis altarium. — Über einige Fledermäuse aus dem nördlichen Thailand mit Bemerkungen zur subgenerischen Zugehörigkeit von Myotis altarium 276 Zingg, P. E.: Search calls of echolocating Nyctalus leisleri and Pipistrellus savii (Mammalia: Chiroptera) recorded in Switzerland. — Suchflugortungslaute von Nyctalus leisleri und Pipistrellus savii (Mammalia: Chiroptera) in der Schweiz 281 Tertenstein, Birgit; Zimmermann, Elke; Rahmann, H.: The development of visual acuity in treeshrews (Tupaia belangeri). - Die Entwicklung der Sehschärfe bei Spitzhörnchen (Tupaia belangeri) 294 aranjon, L.; Cheylan, G.: Mecanismes de coexistence dans une guilde de murides insulaires (Rattus rattus L., Apodemus sylvaticus L. et Mus musculus domesticus Rutty) en Corse: Consequences &Evolutives. — Mechanisms of coexistence in a guild of insular Murids (Rattus rattus L., Apodemus sylvaticus L. and Mus musculus domesticus Rutty) in Corsica: Evolutionary consequences. — Mechanismen der Koexistenz von Insel-Muriden (Rattus rattus L., Apodemus sylvaticus L. und Mus musculus domesticus Rutty) auf Korsika: Evolutionäre Konsequenzen 301 Nissenschaftliche Kurzmitteilung (ierdorf, U.; Kierdorf, H.: Weitgehende Rotation des 4. Prämolaren im Unterkiefer eines Rothirsches (Cervus elaphus L.) und eines Rehbockes (Capreolus capreolus L.) 317 3uchbesprechungen 320 Verlag Paul Parey Hamburg und Berlin HERAUSGEBER/EDENORS P. J. H. van BREE, Amsterdam - W. FIEDLER, Wien - H. Frick, München - W. HERRE, Kiel - H.-G. Krös, Berlin - H.-J. Kunn, Göttingen - E. Kurzer, Tübingen - B. Lanza, Florenz —- J. NIETHAMMER, Bonn — H. ReıcHstein, Kiel - M. Rönrs, Hannover - D. STARcK, Frankfurt a. M. — F. Strauss, Bern -— E. THenıus, Wien - P. Vocer, Lausanne SCHRIETLETTUNG/EDITORTAIOEFERIECR H. SCHLIEMANN, Hamburg —- D. Kruska, Kiel This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con- tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information Die Zeitschrift für Säugetierkunde veröffentlicht Originalarbeiten und wissenschaftliche Kurzmittei- lungen aus dem Gesamtgebiet der Säugetierkunde, Besprechungen der wichtigsten internationalen Literatur sowie dıe Bekanntmachungen der Deutschen Gesellschaft für Säugetierkunde. Verantwort- licher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Harald Schliemann. Zusätzlich erscheint einmal im Jahr ein Heft mit den Abstracts der Vorträge, die auf der jeweiligen Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde gehalten werden. Sie werden als Supplement dem betreffenden Jahrgang der Zeitschrift zugeordnet. Verantwortlich für ihren Inhalt sind ausschließlich die Autoren der Abstracts. Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z. Hd. Prof. Dr. Dieter Kruska, Institut für Haustierkunde, Biologie-Zentrum, Neue Universität, Olshausenstr. 40-60, D-2300 Kiel. Für die Publikation vorgesehene Manuskripte sollen gemäß den „Redaktionellen Richtlinien‘ abgefaßt werden. In ihnen finden sich weitere Hinweise zur Annahme von Manu- skripten, Bedingungen für die Veröffentlichung und die Drucklegung, ferner Richtlinien für die Abfassung eines Abstracts und eine Korrekturzeichentabelle. Die Richtlinien sind auf Anfrage bei der Schriftleitung und dem Verlag erhältlich. Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der Korrekturfahnen erfolgen. Vorbehalt aller Rechte: Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei- ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grund- sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber- rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of the fırst page of an article in this journal indicates the copyright owner’s consent that copies of the article may be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective, or for resale. For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the CCC. Mit einer Beilage des Verlages Paul Parey Fortsetzung 3. Umschlagseite © 1988 Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstr. 12, D-2000 Hamburg 1; Lindenstr. 44-47, D-1000 Berlin 61. — Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., D-2240 Heide/Holst. Z. Säugetierkunde 53 (1988) 257-266 ? © 1988 Verlag Paul Parey, Hamburg und Berlin ea ISSN 0044-3468 SuHSO, Ergebnisse automatischer N Ba an Wochenstubenkolonien der Bechsteinfledermausz, ” (Myotis bechsteini) Von IRMHILD WoLz II. Zoologisches Institut der Universität Erlangen-Nürnberg Eingang des Ms. 9. 6. 1987 Abstract Results of automatically monitoring Bechstein bats’ activities Studied activities of Myotis bechsteini at their roosting sites. Using a newly developed data-acquisition system, the activities of Bechstein bats (Myotis bechsteini) were monitored during 27 observation nights in northern Bavarıa. The following results were obtained: 31-34 min after sunset, one half of a bat colony left for foraging. Bats started to leave the nest boxes at a luminance level of about 14 lux, and 50 % had left by 0.4 lux. The Bechstein bats’ outburst pattern was characterized by departure in groups of several individuals. Both the colony size and environmental factors such as low ambient temperatures influenced the bats’ outburst behavior. Bechstein bats’ activities in front of their roosting nest boxes diminished during the night as the young got older. Bechstein bats returned to a roost 110 min before sunrise. Sixty to forty min before sunrise, the activities peaked in front of the roost; and at 20 mın before sunrise, all bats had returned. Einleitung Als versteckt lebende Baumfledermaus ist die Bechsteinfledermaus (Myotis bechsteini Leisler in Kuhl, 1818) einer Beobachtung bei abendlichem Ausflug und nächtlicher Aktivität schwer zugänglich. Im Bereich des Forstamtes Ebrach im nördlichen Steigerwald (Nord- bayern) nimmt Myotis bechsteini Nistkästen als Quartier an. Charakterisiert durch häufigen Quartierwechsel (Woız 1986) besiedeln die Fledermäuse ab Ende Juli eine Reihe von Vogelnistkästen im Forstrevier, bis sie Ende September wieder verschwinden und sich entweder in unzugängliche Baumquartiere zurückziehen oder das Gebiet verlassen. Mit Hilfe von Infrarotlichtschranken-Ereignisspeichern - entwickelt am II. Zoologischen Insti- tut der Universität Erlangen — konnten abendlicher Ausflug, nächtliche Aktivität am Nistkasten und morgendliche Rückkehr aufgezeichnet werden. Dieses automatische Daten- erfassungssystem ermöglicht mit vertretbarem Arbeitsaufwand Registrierungen der Fleder- mausaktivität über längere Zeit. Material und Methode Drei verschiedene Wochenstubenkolonien der Bechsteinfledermaus, die sich - an den Ringnummern erkenntlich —- nicht mischen, besiedeln den Winkelhofer Forst im nördlichen Steigerwald, einen Laubmischwald mit abwechslungsreicher Waldstruktur. Einzelne Gruppen aus diesen Kolonien - in der Regel adulte Weibchen mit Jungtieren - suchen 30 von insgesamt 575 vorhandenen Nistkästen bevorzugt auf. An 9 dieser Nistkästen konnten in den Jahren 1984 bis 1986 mit Hilfe einer Infrarot- Lichtschranke 27 Ausflugsbeobachtungen durchgeführt werden, wobei insgesamt 636 Registrierungen beim abendlichen Ausflug aufgezeichnet wurden. Der Schwerpunkt lag auf der Überwachung des Nistkastens, der von allen im Forstamt Ebrach angebrachten Nisthilfen von Myotis bechsteini am häufigsten als Quartier genutzt wurde (15 Ausflugsbeobachtungen mit insgesamt 403 registrierten Individuen). In 10 Fällen wurde zusätzlich zur nächtlichen Aktivität auch die morgendliche Rückkehr U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5305-0257 $ 02.50/0 2958 Irmhild Wolz von Fledermäusen erfaßt. Die Koloniegröße war starken Schwankungen unterworfen und betrug mindestens 6, maximal 51 Individuen; die Beobachtungen beschränkten sich jeweils auf die Monate August und September, außerhalb dieses Zeitraumes waren Bechsteinfledermäuse nur gelegentlich in Nistkästen anzutreffen. Alle Zeitangaben erfolgen in mitteleuropäischer Sommerzeit, MESZ. Der Lichtschranken-Ereignisspeicher Um die Probleme mechanischer Schreiber (begrenzte Zeitauflösung, hoher Papierverbrauch) zu umgehen, wurde für die vorliegenden Untersuchungen am II. Zoologischen Institut der Universität Erlangen-Nürnberg ein Infrarotlichtschranken-Ereignisspeicher (IR-LES) entwickelt, der aus den Modulen: 1. Lichtschranken-Rahmen mit IR-Sende- und Empfangselektronik, 2. Ereignisklassifizie- rungs- und Speichereinheit mit Echtzeituhr sowie 3. Datendisplay und Ausgabeeinheit besteht. Aus Stabilitätsgründen wurden Metallrahmen zur Halterung der Sende- und Empfangsdioden entworfen, welche an die Abmessungen der im Forstamt Ebrach verwendeten Nistkästen angepaßt sind (Bayeri- scher Spitzgiebelkasten der Fa. Grund, Fluglochöffnung oval, 3cm x 5 cm). Diese Konstruktion gewährleistet, daß die Schranke rasch am Nistkasten angebracht und justiert werden kann, ohne daß die Fledermäuse durch Erschütterungen gestört werden. Abb. 1 zeigt die Lage der beiden IR-Strahlen relativ zur Fluglochöffnung. Abb. 1. Lichtschranken-Rahmen am Nistkasten. Sen- der- und Empfängerelektronik („S“, „E“) sind regensi- cher untergebracht. Die Lage beider IR-Strahlen („N“, „F“) zwischen den Empfangsdioden D,, D, und den Sendedioden wird durch die unterbrochenen Linien markiert. Mit den Stellschrauben S; und S> kann der Lichtschranken-Rahmen schnell und geräuschlos am Nistkasten befestigt werden. Der Bereich zwischen „N“-Strahl und Nistkastenfront stellt die Zone dar, in der sıch die Fledermäuse hinter den IR-Strahlen bewe- gen können, ohne registriert zu werden Die Richtungserkennung eines Durchfluges erfolgt durch Vergleich der beiden Empfängersignale, ein ‚Ereignis‘ beginnt mit der Unterbrechung des ‚N‘-Strahls (s. Abb. 1). Aus dem Zustand des nistkastenfernen ‚F‘-Strahls zu den Zeitpunkten ‚N-Strahl-Unterbrechung‘ und ‚N-Strahl-Freigabe‘ ergibt sich die gewünschte Ereignisunterscheidung in ‚Einflug‘ und ‚Ausflug‘. Das Verhalten der Fledermäuse beim Ausflug ins Jagdgebiet - schnelles Abspringen vom Flugloch — erlaubt am Abend die exakte Zählung der Individuen einer Wochenstubenkolonie. Der Einflug ins Quartier während der Nacht und am Morgen erfolgt jedoch erst nach einer Reihe von An- und Abflügen am Nistkasten. Ein- und Ausflugsregistrierungen heben sich bei den ‚Probeanflügen‘ nicht immer auf. Ursache hierfür ist folgendes beobachtete Verhalten: Die Fledermäuse fliegen das Flugloch frontal an, schlüpfen jedoch nicht ein, sondern lassen sich nach unten fallen und fliegen nach hinten ab, ohne die IR-Strahlen noch einmal in Ausflugsrichtung zu durchkreuzen. Diese Strahlenanordnung erlaubt somit die präzise Ausflugszählung sowie die Überwachung der Flugaktivitäten vor dem Nistkasten im Verlauf einer Nacht. Automatische Aktivitätsaufzeichnungen bei Myotis bechsteini 259 Helligkeitsmessungen Zur Helligkeitsmessung während der Aktivitätsaufzeichnungen wurden folgende Geräte verwendet: 1. Gossen, Mod. Mavolux digital (Beleuchtungsstärkemeßßgerät, Messung in Ix), 2. Digital-Luxmeter, IC-Sensor Siemens TFA 1001 (Beleuchtungsstärke-Meßgerät), 3. Li-Cor, Mod. LI-185B, Sensor LI- 200SB (Bestrahlungsstärke-Meßgerät, Messung in W/m?). Die Bestrahlungsstärke spiegelt die gesamte einstrahlende Leistung pro Fläche wider, berücksichtigt also auch alle nicht sichtbaren Anteile ım UV- und IR-Bereich. Die Beleuchtungsstärke gibt nur den sichtbaren Anteil der einfallenden Strahlung an und bewertet das Spektrum mit der international festgelegten Gewichtungskurve V(X), welche die spektrale Empfindlichkeit des menschlichen Auges - und nicht des Fledermausauges - berücksichtigt. Einige Vergleichswerte sollen zur besseren Vorstellung der ın der vorliegenden Untersuchung angegebenen Helligkeiten dienen: Sonnenlicht im Sommer 100 klx, im Winter 10 klx, bedeckter Himmel im Sommer 5 klx bis 20 klx, Nacht bei Vollmond 200 mix, mondlose, klare Nacht 0,3 mlx (KucaLing 1982). Ergebnisse Zeitpunkt und Dauer des abendlichen Austlugs Abb. 2 zeigt alle Ausflugsregistrierungen am beliebtesten Nistkasten der größten Myotis- bechsteini-Kolonie. Dargestellt ist die Summe der Ausflüge in der i-ten min nach Sonnen- untergang (SU), summiert über N Beobachtungsnächte. Die dunklen Säulen geben die Austlugsbeobachtungen von 10 Abenden mit vergleichbaren Witterungsverhältnissen und bei mittleren Koloniegröfßen wieder. Man erkennt, daß die ersten ausfliegenden Fleder- mäuse ab der 13. mın nach Sonnenuntergang registriert wurden, dıe Anzahl der das Quartier verlassenden Tiere schnell zunahm und sich dabei relativ hohe Zahlen austliegen- E A: 289 Ausflüge, N=10 E 20-1 3: 403 Ausflüge, N=15 IN © L ® = 8 nes v0 Abb. 2. Abendlicher Ausflug am be- __ liebtesten Nistkasten der größten = Myotis-bechsteini-Kolonie. Blellegn or Säulen: Zusammenfassung der Aus- N flugsregistrierungen an allen N Beob- achtungsabenden (N=15). Helle 0 a a u ae Säulen erreichen auf Grund der Sum- zu ! 10 20 30 n at au wu mendarstellung mindestens die Höhe za eU) (min) dunkler Säulen, da gilt: ZiW)=& St) =N *Rutt). k=1 27.08.1986 (SU = 20.12) Zı(t): Summe der Ausflüge in der i-ten mın nach Sonnenuntergang ın N Nächten; S,(t;): Ausflüge in der i-ten min nach SU in der k-ten Nacht; Rx(t;): Mittelwert der Ausflüge in der i-ten min nach SU, gemittelt über N Nächte; dunkle Säulen: Summendar- stellung von 10 Ausflugsbeobachtun- gen an Kolonien mittlerer Größe un- ter gleichen Witterungsbedingungen (N = 10, Teilmenge von (B); obere Kurve: Abnahme der Helligkeit vor dem Nistkasten am 27. 8. 1986 bei m N M 20 30 40 5 bewölktem Himmel t — t(SU) (min) () 60 70 260 Irmhild Wolz der Individuen mit jeweils im Vergleich dazu niedrigen Werten abwechselten. Hier wird trotz der Summierung von 10 Beobachtungsabenden das pulkweise Ausfliegen der Bech- steinfledermäuse noch deutlich. 50 % aller Fledermäuse sind im Mittel 31.5 min nach Sonnenuntergang ausgeflogen. Mit hellen Säulen zeigt Abb. 2 die Gesamtheit aller Aus- flugsbeobachtungen an diesem Nistkasten einschließlich der Ausflüge, die vom Normal- verhalten deutlich abwichen (z.B. sehr große oder sehr kleine Fledermausgruppen). Im Vergleich zu der mit dunklen Säulen dargestellten Teilmenge (Beobachtungen bei mittlerer Koloniegröße) verschiebt sich das zeitliche Mittel geringfügig: 34 min nach Sonnenunter- gang haben bei Berücksichtigung aller Austlugsregistrierungen (insgesamt 403) 50 % der Fledermäuse das Quartier verlassen. An 7 von 27 Beobachtungsabenden verließen die ersten Individuen schon vor der 20. min nach SU (frühestens ın der 12. min nach SU) das Quartier. An 15 von 27 Abenden begann der Ausflug der Bechsteinfledermäuse zwischen der 20. und 30. min nach SU. Austlugsbeginn eine halbe Stunde nach Sonnenuntergang oder später war dagegen selten und bis auf eine Ausnahme auf kleine Fledermausgruppen mit weniger als 10 Individuen beschränkt. Mit der Individuenzahl der Fledermausgruppen im Nistkasten stieg sowohl die Dauer des Ausfluges wie auch die Ausflugsrate (d.h. die Zahl der austliegenden Tiere pro Zeiteinheit). Im Falle kleiner Gruppen (ca. 5 Individuen) verließen im Mittel 0,5 Tiere/min das Quartier, die mittlere Ausflugsrate großer Gruppen (ca. 40 Tiere) betrug hingegen 2 Tiere/min. Das Abfliegen der Fledermäuse verlief nicht gleichmäßig; bei mittleren oder großen Gruppen bildeten sich häufig Pulks aus 3 bis 5 Fledermäusen. Diese Tiere flogen in zeitlichen Abständen von wenigen Sekunden aus. Bestimmende Faktoren des Ausflugsbeginns Es ıst bekannt, daß die abendlichen Lichtverhältnisse und das schnelle Absinken der Helligkeit während der Dämmerung eine Schlüsselfunktion bei der Bestimmung des Ausflugsbeginns einer Fledermauskolonie einnehmen (ENGLÄNDER und LAUFENS 1968; DECousseEy und DECouRsEY 1964; SwirFt 1980; VOUTE et al. 1974; u.a.). Abb. 2 zeigt die Abnahme der Helligkeit (obere Kurve) am 27. 8. 1986 (bewölkter Himmel) in der Nähe des Nistkastens, an dem die der Graphik zu Grunde liegenden Ausflugsbeobachtungen durchgeführt wurden. Es zeigt sich an der Summendarstellung, daß ım Mittel 50 % der Fledermäuse den Nistkasten zum Jagdflug verlassen hatten, bevor 33 min nach Sonnenuntergang die Helligkeitswerte unter 0,4 1x absanken. Die früh ausfliegenden Fledermäuse (ab der 13. min nach SU, 14 lx am 27. 8. 1986) bewegten sich ın allen drei Beobachtungsjahren noch im hellen Dämmerlicht und waren mit bloßem Auge gut erkennbar, bis sie beim Abflug ins Jagdgebiet hinter Laubwerk verschwanden. Im Jahr 1986 wurden Helligkeitsmessungen sowohl bei vollständig bedecktem wie auch beı klarem Himmel — mit durchziehenden Wolkenfeldern — durchgeführt. Während der frühen Dämmerung war der Einfluß der Bewölkung auf die gemessenen Helligkeitswerte ausge- prägt beobachtbar. Der Helligkeitsbereich, bei dem Bechsteinfledermäuse austlogen, ist jedoch zumindest für einzelne Individuen sehr groß und umfaßt nach Abb. 2 alle Werte unter 15 lx. Helligkeitsschwankungen zur Ausflugszeit durch aufziehende Wolkentelder oder beginnenden leichten Regen beeinflußten das Verhalten der Fledermäuse nicht signifikant. | Im Gegensatz zu den Beobachtungen, die LaurEns (1972) an einzeln lebenden Männ- chen der Bechsteinfledermaus machte, verhinderten starke Regenschauer zur normalen Ausflugszeit - am 26. 8. 1986 und am 28. 8. 1986 — den Aufbruch einer Wochenstubenko- lonie nicht. Sowohl der Zeitpunkt des Ausflugsbeginns (13. bzw. 28. min nach SU) wie auch die Dauer des Ausflugs (33 Indiv. in 31 min bzw. 34 Indiv. in 37 min) unterschieden sich nicht signifikant von Beobachtungen bei schönem Wetter. Bemerkenswerterweise Registrierungen (Ereignisse / 5min) Registrierungen (Ereignisse / 5min) Automatische Aktivitätsaufzeichnungen bei Myotis bechsteini 261 wies das Ausflugsgeschehen am 26. August allerdings die größte je registrierte Zahl langer Pausen auf (7 Pausen über 3 min). Deutliche Auswirkungen auf das Verhalten der Fledermäuse hatten hingegen niedrige Aufßßentemperaturen, die den Ausflug um bis zu 20 min verzögerten bzw. ganz verhinder- ten. Die Abendtemperaturen —- gemessen um 19 Uhr - nahmen vom 26. 8. 1986 (18°C) bis zum 29. 8. 1986 (10°C) deutlich ab. Drei Fledermäuse aus einer Gruppe von 7 Individuen flogen am 29. 8. stark verspätet aus (48.-54. min), die restlichen Tiere blieben während der gesamten Nacht im Quartier. Dieses Verhalten wurde beı vergleichbaren Temperaturver- hältnissen auch in anderen Jahren beobachtet; teilweise flogen die Fledermäuse auch an zwei aufeinanderfolgenden Nächten nicht zum Jagdflug aus. Ebenfalls Einfluß auf den Ausflugsbeginn scheint neben der Helligkeit und den Wetterbedingungen die Koloniegröße zu haben. Kleine Fledermausgruppen neigten zu spätem Ausflug ab 30 mın nach SU. Auch zeigte die größte beobachtete Kolonie von Myotis bechsteini (51 Indiv.) ungewöhnliches Verhalten mit Ausflugsbeginn in der 36. min nach SU und sehr raschem Ausflug in 12 min. Es konnten jedoch keine statistisch signifikanten Daten ermittelt werden, da derart große Gruppen zu selten zu finden waren. Aufzeichnungen zwischen abendlichem Ausflug und morgendlicher Rückkehr Abb. 3 (a-c) zeigt das Verhalten einer Wochenstubenkolonie in drei Nächten des Jahres 1984. Der Nistkasten wurde im August 1984 15 Tage lang von den Fledermäusen ununterbrochen als Quartier benutzt. So war es möglich, Ausflug und Rückkehr der Bechsteinfledermäuse mehrmals hintereinander zu dokumentieren. Es zeigte sich, daß die 14./15.08.1984 09./10.08.1984 30) | Aust iug [ES Einflug Een a u BER Aust us | Einf; Pin m oO m De} [——] ii nal) o- [=] I =] I nm oO Registrierungen (Ereignisse / 5min) SU = 20.36 SA = 6.05 20.00 20.30 22.30 25.00 23.30 0.00 0.30 1.00 1.30 2.00 2.30 3.00 3.30 4,00 4.50 9.00 5.50 6.00 20.00 20.30 21.00 21.30 22.00 22.30 23.00 23.50 0.00 0.30 1.00 1.30 2.00 2.30 3.00 3.30 4.00 4.30 3.00 5.30 6.00 Uhrzeit Uhrzeit nr 19./20.08.1984 45 Individuen C 20 Abb. 3. Nächtliche Aktivität einer Kolonie von Myotıs bechsteini ım August 1984. a: In der Nacht vom 9. auf den 10. August 1984 herrscht vor dem Nistkasten während der m ganzen Nacht heftige Aktivität durch an- und abtliegende Fledermäuse; b: In der Nacht vom 14. auf den 15. August SU = 20.27 sA=6.13 1984 zeichneten sich drei Aktivitätsmaxima ab; c: Zwi- -20 schen dem Ausflug am 19. August 1984 und dem Einflug : am 20. August 1984 wurden trotz großer Individuenzahl der Kolonie (45 Tiere) nur wenige Ereignisse während der Uhrzeit Nacht registriert 20.00 20.30 21.00 21.30 22.00 22.30 23.00 23.30 0.00 0.30 1.00 1.30 2.00 2.30 3.00 3.30 4,00 4.30 3.00 5.30 6.00 262 Irmhild Wolz nächtliche Aktivität am Nistkasten vom 9./10. August bis zum 19./20. August deutlich abnahm. Waren es ın der ersten Beobachtungsnacht bei 35 Individuen noch 588 Ereignisse, verringerte sich diese Zahl in der Nacht vom 19. auf den 20. August bei einer Koloniegröße von 45 Individuen abends bzw. 37 Individuen morgens auf 83 Ereignisse. Vom 9. auf den 10. 8. 84 herrschte während der gesamten Nacht starke Aktivität vor dem Nistkasten, vom 14. auf den 15.8. 84 waren noch drei Aktivitätsmaxima erkennbar, in der letzten Beobachtungsnacht registrierte der LES bis zum Eintlug ab 5 Uhr nur vereinzelt Ereig- nisse. Am 24. 8. 1984 kehrten die Fledermäuse am Morgen nicht mehr in ihr Quartier zurück, einige Individuen flogen den Nistkasten ab 4 Uhr jedoch mehrmals an, die letzte Fledermaus wurde dabei 30 min vor Sonnenaufgang registriert. In zehn Nächten wurden Beobachtungen an quartierwechselnden Kolonien aufgezeichnet, die Zahl der Registrie- rungen am bisherigen Tagesquartier blieb in der Regel kleiner als 10 - nur in einer Nacht kehrte keine Fledermaus zurück. Die meisten Tiere suchten den verlassenen Nistkasten erst zur normalen morgendlichen Einflugszeit auf (in 6 von 9 Fällen), flogen aber wieder ab. Die späteste Fledermaus erschien dabei — offensichtlich auf der Suche nach ihrer Kolonie - 5 min vor Sonnenaufgang. Morgendlicher Einflug der Tiere In 10 Nächten (Aug. 84, Aug./Sept. 85, Aug. 86) gelang es, auch das morgendliche Rückkehrverhalten der Bechsteintledermäuse zu dokumentieren. Abb. 4 zeigt die Zusam- menfassung von 10 Einflugsbeobachtungen und die gemessenen Helligkeitswerte kurz vor Sonnenaufgang (SA). 110 min vor SA setzte der Einflug der Fledermäuse ein; 60-40 min vor SA herrschte die größte Aktivität vor dem Nistkasten, ab 30 min vor SA fanden nur noch einzelne Einflüge statt. Die Aufzeichnungen der zwischen drei und zwei Stunden vor Sonnenaufgang zurückkehrenden Fledermäuse stammen von Beobachtungen in der jeweils ersten Augusthälfte und sınd vermutlich auf Aktivitäten von Jungtieren zurückzuführen. Je später die Aufzeichnungen im August erfolgten, desto deutlicher zeichnete sich der morgendliche Einflug zwischen der 80. und 30. min vor SA ab (s. Abb. 3). Die Helligkeitswerte während des Einflugs am 28. 8. 86 zeigt die obere Kurve in Abb. 4. In mehreren Fällen wurde die Rückkehr der Fledermäuse (mit Nachtsichtgerät bzw. bei ausreichender Helligkeit mit blofßem Auge) beobachtet und gleichzeitig die Ereignisse mit dem LES dokumentiert. Dies geschah z.B. am 29. 8. 84, als eine kleine Gruppe von 10 Bechsteinfledermäusen in ıhr vorheriges Tagesquartier zurückkehrte. Um 5 Uhr - bei noch völliger Dunkelheit — umtlogen bereits einzelne Fledermäuse in einer Höhe von 2 bis 4m das Quartier, verschwanden häufig minutenlang, um mit zunehmender Helligkeit den Nistkasten immer näher zu umrunden. Dann begannen sie, das Flugloch seitlich und von vorn anzufliegen. Einzelne Individuen klammerten sich am Nistkasten vor dem Flugloch fest, ließen sich fallen und flogen wieder ab. Der LES zeichnete ab 5.15 Uhr die ersten Ereignisse auf, der Einflug von 10 Fledermäusen ergab insgesamt 32 Registrierungen, der letzte Einflug erfolgte um 6.07 Uhr (Sonnenaufgang am 29. 8. 1984: 6.26 Uhr). Während der Endphase des Einflugs ab 5.45 Uhr wurde als Maximum auf 13 Anflüge an das Flugloch nur 1 Einflug beobachtet. Diskussion In der bisherigen Literatur zum Themenkreis ‚Ausflugszeit‘, ‚Austlugsverhalten‘ bzw. ‚Aktivitätsperiodik der Fledermäuse‘ finden sich nur vereinzelt Hinweise auf die Bech- steinfledermaus (Myotis bechsteini) (Kos 1959; Laurens 1972; Laurens 1973b). Es handelt sich dabei um Beobachtungen an Einzeltieren und nicht wie in der vorliegenden Untersuchung an Wochenstubenkolonien. Die Meinung, daß Bechsteintledermäuse erst Automatische Aktivitätsaufzeichnungen bei Myotis bechsteini 263 BE Einiiug ® Ausflug En. oO = nr 10 Beobachtungsnächte BI L © eg! 0 = Tr er 1 On u NAAR ij "m \ ! -20 Abb. 4. Zusammenfassung von 10 200 BOE 160 te 170710022 B 02600240 5220 52,105:520 morgendlichen Einflugsbeobachtun- t — t(SA) (min) gen am beliebtesten Nistkasten der größten Myotis-bechsteini-Kolonie. 28.08.1986 (SA = 6.24) 60 bis 40 min vor Sonnenaufgang herrscht die größte Aktivität vor dem Nistkasten. Die Registrierungen zwi- schen 3 und 2 h vor Sonnenaufgang % sind auf Rückflüge von Jungtieren < während der ersten Augusthälfte zu- rückzuführen. Zg(t)):; Summe der = Einflugsregistrierungen in der ı-ten Y min vor Sonnenaufgang ın 10 Näch- 6) ten (zur Definition von Zr(t;) vgl. = Abb.4, N = 10). Negative Werte: ® Ausflugsregistrierungen. Die obere - Kurve zeigt die Zunahme der Hellig- keit vor dem Nistkasten am 28. 8. 1986 = BD, 60m a0 120 (oDrR Bo 0) a0 t — t(SA) (min) nach Einbruch der Dunkelheit austliegen (TAAKE 1985), wurde durch das Verhalten von Myotis bechsteini ım Steigerwald nicht bestätigt. Der abendliche Ausflug erstreckte sich — abhängig von der Koloniegröße - teilweise über den gesamten Bereich der Dämmerung, die ersten austliegenden Individuen waren bei hohen Helligkeitswerten mit bloßem Auge gut erkennbar, während die letzten Fleder- mäuse ın der Dunkelheit nur mit Hilfe automatischer Registrierung zu erfassen waren. Um die Austlugszeit der Bechsteinfledermäuse mit der anderer Arten zu vergleichen, sind Angaben zur Ausflugszeit in Minuten nach Sonnenuntergang nicht geeignet, wenn nicht gleichzeitig Helligkeitsmessungen durchgeführt werden. So geben viele Autoren Zeiten zwischen 20 und 35 min nach Sonnenuntergang für die ersten ausfliegenden Tiere an: Zwergfledermaus, nördliches Schottland, 35 min: Swırt (1980); Mausohr, FRG, 30 min: DECoursey und DECoussey (1964); Abendsegler in Dänemark, 20-30 min: v. HEERDT und SLUITER (1965); Teichfledermaus, >30 min, Niederlande: VOOTE et al. (1974). Erst Helligkeitsmessungen machen die Unterschiede zwischen den Fledermausarten deutlich: Zwergfledermäuse fliegen bei Werten zwischen 15-35 lx aus (SwıFt 1980), Mausohren bei 20-100 mlx (DECoursey und DECoursEY 1964), Teichtledermäuse zumeist bei Werten unter 350 uW/m? (VoOTE et al. 1974). Die niedrigsten Helligkeitswerte für ausfliegende Tiere wurden für Myotis nattereri ermittelt, die nach ENGLÄNDER und LAUFENS (1968) Ende August und ım September erst 10-30 min nach Absınken der Helligkeit auf 1 mlx ihre Quartiere verlassen. Myotis bechsteini fliegt nach den Ergebnissen der vorliegenden Untersuchung in der Regel bei Helligkeitswerten unter 14 lx aus, 50 % der Individuen verlassen das Quartier nach Absinken der Helligkeit unter 0,4 Ix. Die Austlugswerte der Bechsteinfledermäuse sind daher am ehesten mit denen des Mausohrs zu vergleichen und 264 Irmhild Wolz liegen noch deutlich vor der Austlugszeit der Fransenfledermäuse (Myotis nattereri), die ihr Quartier erst nach Einbruch der Dunkelheit verlassen. Bei kleinen Fledermauskolonien begann der Ausflug relativ spät. Die Ursache dafür ist u.U. darin zu sehen, daf sich wenige, im Nistkasten einzeln hängende Tiere beim abendlichen Erwachen gegenseitig nur geringfügig stören, sich die Unruhe der ersten erwachenden Fledermaus also nicht oder nur wenig auf die anderen Individuen der Gruppe überträgt. In einem voll besetzten Nistkasten dagegen wird das erste unruhig werdende Tier auch die anderen Fledermäuse beeinflussen und in Kürze die gesamte Kolonie aus dem Tagesschlaf wecken, unabhängig von den verschieden synchronisierten inneren Uhren der Einzeltiere. In kurzer Folge hintereinander austliegende Tiere (‚Pulks‘) wechseln sich dann mit einzeln abfliegenden Fledermäusen ab. Swırr (1980) sieht dafür drei mögliche Ursa- chen: 1. Pulkweises Austliegen ist das Ergebnis sozialer Stimulation, das Zeigen eines Verhaltensmusters durch ein Mitglied der Kolonie wirkt als Trigger für dasselbe Verhalten bei anderen Individuen und die Fledermäuse erregen sich so jeweils gruppenweise in Flugbereitschaft; 2. pulkweises Ausfliegen dient als Mittel zur Verwirrung von Freßfein- den; 3. die Fledermäuse jagen ın Gruppen und warten am Ausflugloch auf andere zum Jagdflug bereite Individuen. Da die Bechsteinfledermäuse auch beim pulkweisen Ausflie- gen das Quartier auf Grund der kleinen Fluglochöffnung immer einzeln verlassen müssen, wenn auch in kurzen Abständen von wenigen Sekunden, so scheint dieses Verhalten zur Verwirrung eines Freßfeindes nicht geeignet, da die Sicherheit einer ausfliegenden Fleder- maus durch ein in kurzem Abstand voraustliegendes Tier nicht erhöht wird. Ebenso ist unbekannt, ob die Bechsteinfledermäuse in Gruppen jagen; nach den bisherigen Beobach- tungen ist dies eher unwahrscheinlich (unterschiedliche Abflugrichtung der Tiere vom Nistkasten). Wenn das pulkweise Austliegen auf soziale Stimulation zurückgeführt werden kann, so sollten die Fledermäuse großer Kolonien deutlicher und ausgeprägter zur Pulkbildung neigen. Dies ıst bei den Bechsteinfledermäusen auch tatsächlich der Fall. Während bei großen Wochenstubenkolonien Pulks austliegender Fledermäuse häufig 4 bis 6 (max. 9) Individuen umfaßten, verließen beim abendlichen Austlug kleiner Kolonien (<15 Indiv.) viele Fledermäuse das Quartier allein; es bildeten sich zwischendurch nur kleine ‚Pulks‘ von 2 bis 3 Individuen. Die Beobachtungen zeigten, daß ungünstiges Wetter, vor allem bei extremen Bedingun- gen, Einfluß auf den Ausflugsbeginn hat. Temperatureinbrüche verzögerten oder verhin- derten den Ausflug der Bechsteinfledermäuse. Die von LAaurens (1973b) bei tiefen Nachttemperaturen beobachtete deutliche Aktivitätsverkürzung bei Myotis bechsteini wird mit der vorliegenden Untersuchung bestätigt. Starke Regenfälle während der Austlugszeit zeigten nur in einem Fall Auswirkungen auf das Austlugsgeschehen: es bildeten sich vermehrt längere Pausen zwischen dem Ausflug einzelner Pulks. Ähnliche Reaktionen auf ungünstige Witterungsbedingungen wurden z.B. von Enc- LÄNDER und LAUFENS (1968) für Myotis nattererı, von BÖHME und NATUSCHKE (1967) sowie von GEBHARD und OTT (1985) für Myotis myotis und von v. HEERDT und SLUITER (1965) für Nyctalus noctula dargestellt. Die nächtliche Aktivität vor dem Quartier ging mit zunehmendem Alter der Jungtiere bemerkenswert deutlich zurück. Da sendertragende adulte Bechsteinfledermäuse in der zweiten Augusthälfte während der Nacht das Tagesquartier nicht aufsuchten (Woız 1986), sind die hohen Aktivitätswerte zwischen abendlichem Ausflug und morgendlicher Rück- kehr vermutlich zum großen Teil auf Jungtiere zurückzuführen. Fransenfledermäuse zeigen nach Laurens (1973a) ein den Bechsteinfledermäusen sehr ähnliches Verhalten mit einer bis Mitte August zunächst größer, dann allmählich kleiner werdenden Zahl von Individuen, die im Laufe der Nacht ins Quartier zurückkehren. Ab Mitte August bis Ende September werden auch bei Myotis nattereri nur wenige nächtliche Rückflüge ins Quartier aufgezeichnet. Automatische Aktivitätsanfzeichnungen bei Myotis bechsteini 265 In den Morgenstunden kehrten die Bechsteinfledermäuse zum Tagesquartier zurück, um es gruppenweise zu umfliegen. Durch dieses Verhalten lernen die Fledermäuse - vor allem die Jungtiere - Lage und Umgebung des Tagesquartiers genau kennen. Während der Zeit, in der die Tiere den Nistkasten noch in weiten Kreisen umflogen und minutenlang verschwanden, wurden vermutlich andere Ruheplätze überprüft. Zu diesem Zeitpunkt fällt dann die Entscheidung für das zukünftige Tagesquartier. Je mehr Individuen eine Gruppe bilden, desto mehr Information steht zur Auswahl eines optimal geeigneten Quartiers zur Verfügung. 60-40 min vor Sonnenaufgang herrschte beim morgendlichen Finflug die größte Aktivität vor dem Tagesquartier. Die Helligkeitswerte beim Einflug lagen dabei deutlich unter denen des Ausflugs, beim Erreichen des Helligkeitswertes von 1 Ix (30 min vor SA) waren nahezu alle Fledermäuse ins Tagesquartier eingeflogen. Dies steht im Einklang mit Laurens (1972), der bei einzeln lebenden Männchen der Bechsteinfledermaus ebenfalls abends größere Flughelligkeiten als beim morgendlichen Einflug feststellte. Die Rückkehr der Tiere - nach seinen Beobachtungen zwischen der 60. und 20. min vor Sonnenaufgang — liegt im gleichen Zeitraum, der in der vorliegenden Untersuchung für Wochenstubenkolo- nıen der Bechsteinfledermaus festgestellt wurde. Danksagung Mein besonderer Dank gilt Herrn Forstdirektor Dr. SPERBER und seinen Mitarbeitern im Forstamt Ebrach für ihre Unterstützung der Freilandarbeiten, Herrn G. Schtarp für das Überlassen von Beringungsunterlagen sowie Herrn Prof. Dr. ©. v. HELVERSEN für kritische Durchsicht des Manu- skriptes und hilfreiche Diskussionen. Der Fa. Gossen, Erlangen, gebührt mein Dank für die freundliche Überlassung von Helligkeitsmeßgeräten. Zusammenfassung Zur Registrierung der Aktivitäten von nistkastenbewohnenden Kolonien der Bechsteinfledermaus (Myotıs bechsteini) wurde ein Infrarotlichtschranken-Ereignisspeicher entwickelt. Er besteht aus einem Lichtschrankenrahmen mit IR-Sender- und Empfängerelektronik, einer Auswerte- und Spei- chereinheit mit Echtzeituhr sowie Datendisplay und Ausgabesteuerung für Drucker oder Datenträ- ger. Die Aufzeichnungen der Aktivitäten von Wochenstubenkolonien der Bechsteinfledermaus in 27 Beobachtungsnächten erbrachten folgende Resultate: 31-34 min nach Sonnenuntergang ist die Hälfte der Bechsteinfledermäuse ausgeflogen. Die Helligkeitswerte beim Ausflug der ersten Hälfte aller Tiere liegen zwischen 14 Ix und 0,4 Ix. 50 % der Individuen fliegen bei Helligkeiten unter 0,4 Ix aus. Die Fledermäuse verlassen das Quartier am Abend in charakteristischer Weise: „Pulkweises Ausfliegen“. Sowohl die Koloniegröße wie auch extreme Wetterbedingungen (z.B. niedrige Temperaturen) beeinflussen Beginn und Verlauf des Ausflugs. Die nächtliche Aktivität vor dem Nistkasten nimmt mit dem Alterwerden der flugfähigen Jungtiere von Mitte August bis Anfang September deutlich ab. 110 mın vor Sonnenaufgang beginnt der morgendliche Einflug der Fledermäuse. 60-40 min vor Sonnenaufgang herrscht dıe größte Aktivität vor dem Nistkasten, ca. 20 min vor Sonnenaufgang ist der Einflug beendet. Die Helligkeits- werte beim Finflug liegen unter denen des Ausflugs. Die an den Wochenstubenkolonien der Bechsteinfledermäuse gewonnenen Ergebnisse werden diskutiert und mit den Beobachtungen anderer Autoren verglichen. Literatur BÖHME, W.; NATUSCHKE, G. (1967): Untersuchung der Jagdflugaktivität freilebender Fledermäuse ın Wochenstuben mit Hilfe einer doppelseitigen Lichtschranke und einige Ergebnisse an Myotıs myotis (Borkhausen, 1797) und Myotis nattereri (Kuhl, 1818). Säugetierkundl. Mitt. 15, 129-138. DECoussEY, G.; DECoussey, P. J. (1964): Adaptive Aspects of Activity Rhythms in Bats. Biol. bull. mar. biol. Lab., Woods Hole 126, 14-27. ENGLÄNDER, H.; LAurEns, G. (1968): Aktivitätsuntersuchungen bei Fransenfledermäusen (Myotis nattereri). Experientia 24, 618-619. GEBHARD, J.; OTT, M. (1985): Etho-ökologische Beobachtungen an einer Wochenstube von Myotıs myotis (Borkh., 1797) bei Zwingen (Kanton Bern, Schweiz). Mitt. Naturforsch. Ges. Bern N.F. 42, 129-144. 266 Irmhild Wolz HEERDT, P. F. v.; SLUITER, J. W. (1965): Notes on the distribution and behaviour of the Noctule bat (Nyctalus noctula) ın the Netherlands. Mammalıa 29, 463-477. Kos, A. (1959): Ein Registrierapparat für Fledermäuse und einige biologische Ergebnisse. Zool. Anz. 163, 135-141. Kuchuing, H. (1982): Taschenbuch der Physik. 4. Aufl. Thun und Frankfurt/M.: Verlag Harri Deutsch. LAUFENS, G. (1972): Freilanduntersuchungen zur Aktivitätsperiodik dunkelaktiver Säuger. Inaug.- Diss., Univ. Köln. — (1973a): Beiträge zur Biologie der Fransenfledermäuse (Myotis nattereri Kuhl, 1818). Z. Säugetier- kunde 38, 1-14. — (1973b): Einfluß der Außentemperaturen auf die Aktivitätsperiodik der Fransen- und Bechstein- fledermäuse (Myotis nattereri Kuhl, 1818 und Myotis bechsteini Leisler, 1818). Period. biol. 75, 145-152. Swirt, $S. M. (1980): Activity patterns of Pipistrelle bats (Pipistrellus pipistrellus) in north east Scotland. J. Zool., Lond., 190, 285-295. TAAKE, K.-H. (1985): Einige verhaltensökologische Aspekte der Räuber-Beute-Beziehungen europäi- scher Fledermäuse (Chiroptera). Z. Säugetierkunde 50, 202-208. VoÜTE, A. M.; SLUITER, J. W.; GRIMM, M. P. (1974): The influence of the Natural Light-Dark Cycle on the Activity Rhythm of Pond Bats (Myotis dasycneme Boie, 1825) during Summer. Oecologia 17, 221-243. Worz, I. (1986): Wochenstuben-Quartierwechsel bei der Bechsteinfledermaus. Z. Säugetierkunde 51, 65-74. Anschrift der Verfasserin: Dipl.-Biol. IRmHıLD WoLz, Institut für Zoologie II, Universität Erlangen- Nürnberg, Staudtstr. 5, D-8520 Erlangen Z. Säugetierkunde 53 (1988) 267-275 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Distribution and biometry of Sorex granarius (Miller, 1910) (Soricinae: Insectivora) By J. GIsBERT, M. J. LÖöPEZ-FUSTER, Rosa GARCIA-PEREA and ]. VENTURA Unidad de Zoologia Aplıcada, Madrid and Facultat de Biologia, Universitat de Barcelona Receipt of Ms. 18. 5. 1987 Abstract The area of distribution of Sorex granarius ıs confined to the Iberian Peninsula, including Galicia, the northern half of Portugal and the Central System. This species occupies forestal and supraforestal biotopes, with an altitudinal distribution from sea level to 2000 m.a.s.l. The interval of altitudes where this shrew ıs found is smaller and higher in the Central System (500-2000 m.a.s.l.). Sorex granarius presents marked craniometric homogeneity, although there is a decremental tendency in size from north to south, the largest specimens occurring in Galicıa. Introduction In the Iberian Peninsula, the shrews of the genus Sorex with sexual trıvalent (araneus- articus group, MEyLAN and HaAusser 1973, araneus group, HAussEr 1976) are represented by three karyologically and biochemically well-characterized species: Sorex araneus Linnaeus, 1758; Sorex coronatus Millet, 1828; and Sorex granarius (MiLLER 1910). Sorex granarius was described by MILLER (1910) as a subspecies of Sorex aranens (S. a. granarius, terra typica La Granja, Segovıa). Nonetheless, as a result of the karyological and morphological analysıs by Hausser et al. (1975), the form granarins was raised to species rank. In 1914, CABRERA attrıbutes a specimen from Vilaboa (La Coruna) to this form and postulates that the distribution of granarius ıs the central and northwestern Iberian Peninsula. Later, various authors report the species in Galicia (HEIM DE Bausac and DE BEAUFORT 1969; NoRES 1979; LOPEZ-FUSTER 1983), the Central System (Hausser et al. 1975; Campos 1977; CATZEFLIS et al. 1982; ArEnAs 1983; CATZEFLIS 1984; HAusser 1984) and Portugal (NIETHAMMER 1970; MADUREIRA and MAGALHAES 1980); MADUREIRA and RAMALINHO 1981; RAMALINHO, ın press). The references by ALmacA (1968) and GARZON- HEyDT et al. (1971) to Sorex araneus in Portugal and the Central System, respectively, ın reality correspond to Sorex granarıns. In contrast, the specimens from Villarreal (Alava) that Marec and STORcH (1964) consider Sorex granarius should be referred to Sorex coronatus (HAUSSER et al. 1975). As for Sorex from northern Burgos (Sedano), bibliographic findings are contradictory. According to NIETHAMMER (1956), these specimens pertain to Sorex granarius in the light of their coloring and dimensions, while GARZON-HEYDT et al. (1971) assign them to Sorex araneus (read coronatus). On the basıs of Hausser’s (1984) multivariate analysis, the specimens from Sedano are Sorex granarıns. Nonetheless, Hausser indicates that this diagnosis must be erroneous according to the geographical distribution of Sorex granarins and Sorex coronatus. In the map of distribution described by Hausser et al. (1985), the specific affıliation of these specimens is not mentioned. The objective of our paper was to establish the morphometric features of Sorex granarius and offer our biogeographical findings. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5305-0267 $ 02.50/0 268 J. Gisbert, M. J. Lopez-Fuster, Rosa Garcia-Perea and ]J. Ventura Material and methods Material analysed Two hundred thirty-five specimens were studied. The material came from the scientific collections of: the Unidad de Zoologıa Aplicada, deposited at El Encin, Alcalä de Henares; the Department of Vertebrates of the School of Biology, Universidad Central de Barcelona; and the Department of Vertebrates of the School of Biology, Universidad de Salamanca. The specific determination was realized on the basis of the cranial criteria proposed by MILLER (1912) and Hausser et al. (1975). For the captured material, also was examined the coloration and body measurements (see CABRERA 1914). Specimens captured: Madrid: Alto de Guarramillas, 1. Segovia: Puerto de Fuenfria, 1. Avila: Laguna de El Barco, 1. Leön: Lago de la Bana, 1. La Coruna: Vallegestoso, 1; Ferrol, 1. Lugo: San @iprian,e2. Owl pellet material: Segovia: Grado del Pico, 1 skull and 2 mandibles; Riofrio de Rıaza, 9 skulls and 9 mandibles. Avila: El Barco de Avila, 75 skulls and 54 mandibles. Cäceres: Herväs, 3 skulls and 3 mandibles; Banos de Montemayor, 1 skull and 1 mandible; Acebo, 1 skull and 1 mandible. Salamanca: Villasrubias, 1 skull and 1 mandible; Candelarıo, 16 skulls and 13 mandibles. La Coruna: Arines, 1 skull and 1 mandible; Brandomil, 6 skulls and 3 mandibles; Puebla de Caraminal, 2 mandibles; Monttero, 1 skull and 2 mandibles; Pontedeume, 7 skulls and 2 mandibles. Orense: Castrocaldelas, 5 skulls and 6 mandibles. Body, skull and jaw measurements We used the following measurements: CC = head + body length. C = tail length. P = hind foot length. O = ear length. Peso = weight in grams. LCI = condyle-incisor length. LCB = condyle-basal length. LR = rostral length. LCC = length of skull case. LSB = staphylion-basion length. AIO = interorbital width. AZ = zygomatic width. ACC = skull case width. SDS = length of upper dental series. P-M? = P*-M? length. « = labial length of the mandible. ß = length of articular process. y = inclination of the coronoid process. ö = mandibular foramen. LIA = incisor-angle length. LM = mandibular length. LFT = length of internal temporal fossa. SDI = length of lower dental series. M-M; = M,-M; length. LM; = maximum M; length, LA = articular length of mandible. HC = coronoid height (Fig. 1). The cranıal measurements and mandibular LA and HC were realized with a Mitutoyo calıper, + 0.02 mm precision. For the other mandibular measurements, a “Reichert Mak MS” stereomicro- scope ocular micrometer was used. The precision of this instrument, by lens, was: 1:1 lens, + 0.065 mm («, LIA, EM, SDI, M, M5); 1:4 lens, #0.015’mm(ß, vJEERSENDESIEHOBTEnS + 0.006 mm (ö). We used the method described by Sans-Coma (1979) for the measurements with stereomicroscope, which ıs inspired by that proposed by Hausser and JaAmMmoT (1974), based on use of the “bloc comparateur optique” conceived by JAMMoT (1973). Biomathematical analyses Data were processed with an IBM 3083/XE01 computer of the Calculation Center of the University of Barcelona. Sample means were compared by variance analysis - ANOVA - and paired samples by the Tukey method. The degree of intersample affinity for each varıable was evaluated by the Student- Newman-Keuls test (Dixon 1983: BMDP program p7D). Results and conclusions Distribution and habitat Sorex granarins ıs distributed throughout the Central System, from Sierra de Ayllon (Spain) to Serra da Estrela (Portugal), extending throughout Portugal from north of the “Tajo” River to Galicıa (Fig. 2). | In the Central System, the species ıs located at altitudes between 500 and 2000 meters, occupying the supra- and oromediterranean bioclimatic levels (Rıvas-MArTINEZ 1981, 1983). In this area, where captures were scant, Sorex granarius was found especially in forest biotopes: woods of Fagus sylvatica (Cantalojas), Pinus silvestris (Balsain, Hoyos del Espino), Quercus pyrenaica (Candelario) and Quercus rotundıfolia (La Maya). Sorex granarins also occupies cultivated lands that replace authocthonous forests, like green Distribution and biometry of Sorex granarius 269 AIO ov Fig. 1. Cranial and mandibular measurements of Sorex granarius gardens (El Barco de Avıla) and forests of Castanea sativa (Herväs) and Pinus pinaster (Villasrubias). Above the forest level, the species was captured in granite rock falls adjacent to areas of Nardus stricta (Laguna de El Barco, Galayos). From the western extreme of the Central System, the distribution of Sorex granarius is at progressively lower altitudes, appearing at the mesomediterranean level and reaching the coast of Portugal and Gadlıcia. According to MADUREIRA and RAMALINHO (1981), the species is present in Portugal ın almost all the biotopes with a bushy stratum, being more abundant in densely vegetated, relatively humid terrain. In northern Portugal, Sorex granarins introduces itself into the eurosiberian region, penetrating in Galıcia, where it occupies the Quercus pyrenaica woods and areas where the natural forest has been replaced by Eucaliptus and Pinus pinaster plantations. Generally speaking, the territories occupied by Sorex granarius are characterized by mean annual temperatures of 3-15 °C with cold to extremely cold winters and mean annual precipitation in excess of 600 mm. J. Gisbert, M. J. Löpez-Fuster, Rosa Garcia-Perea and J. Ventura 270 w 0001 0} Jaddn sapnyıyyyf 18497 ueauelssyıpeweuidng jo 4apıog - - - - 19A97] LeauUeLJ 1 21PEWOWIBL Jo 19pJog -'- "— uolbsy uelusgısoing Jo 1ap10g AB hs x r “ar Ei 3% Das klassısche Werk über dıe Anatomie der Wirbeltiere Dieses anschauliche Werk ist vor allem für Studenten der Biologie, der Medizin und verwandter Studienrichtungen gedacht; es gibt aber auch den Spezialisten auf engeren Forschungsgebieten der Biologie zuverlässige Grundlagen für ihre wissenschaftliche Arbeit. Darüber hinaus wendet es sich an jeden, der am Beispiel der Wirbeltieranatomie die wundervolle Mannigfaltigkeit der Organismen zu verstehen sucht. %* Alfred Sherwood Romer/Thomas S. Parsons: Vergleichende Anatomie der Wirbeltiere. Aus dem Amerikanischen übersetzt und bearbeitet von Hans Frick. Mit einem Geleitwort von Dietrich Starck. 5., neubearbeitete und erweiterte Auflage. 1983. 639 Seiten mit 1310 Einzeldarstellungen, davon 123 farbig, in 435 Abbildungen. Gebunden 48,- DM Verlag Paul Parey Spitalerstraße 12 2000 Hamburg 1 PAJL PREV Distribution and biometry of Sorex granarins 271 Biometry The origin of the material, predominantly from owl pellets, precludes our establishing the somatometric and biological characteristics of Sorex granarins. Nonetheless, since the bibliographic data on these aspects are restricted to those communicated by MILLER (1912), we consider it interesting to offer the body measurements obtained in 10 specimens from diverse localıties (Table 1). Table 1. Body measurements of Sorex granarius Locality Date Balsain (Segovia) 2041977, Alto de Guarramillas (Segovia) Laguna de El Barco (Avıla) Gontän-Quende (Lugo) Vilas, S. Ciprian (Lugo) S. Ciprian (Lugo) Ferrol (Coruna) Ferrol (Coruna) Vallegestoso (Coruna) El Barco de Aviıla (Aviıla) * = weight corresponding to a pregnant female Moreover, in view of the fragmentary nature of cranıometric information, we list ın Table 2 the values of cranıal varıables obtained in our specimens from Galıcıa (the data from Gredos and Guadarrama are insufficient). Our findings concur with those communi- Fıg. 2. Distribution of Sorex granarius ın the Iberian Peninsula. Black circles: own data. White cırcles: references in literature. 1: Grado del Pico (Segovia), 2: Cantalojas (Guadalajara), 3: Riofrio de Rıaza (Segovia). 4: Cerezo de Arrıba (Segovia), 5: Pto. de Somosierra (Madrid), 6. Rascafria (CATZEFLIS et al. 1982). 7: Alto de Guarramillas (Madrid), 8: Balsain (Garzon et al. 1971), 9: La Granja (MILLER 1910). 10. Cercedilla (Madrid), 11: Pto. de Fuenfria (Segovia), 12: Galayos (Avila), 13: Barajas (Avila), 14: Hoyos del Espino (Avila), 15: Navalperal de Tormes (Avila), 16: Piedrahita (HAusser 1984), 17: Santiago del Collado (Avila), 18: El Barco de Aviıla (Avıla), 19: Laguna de El Barco (Aviıla), 20: Candelario (Hausser et al. 1975), 21: La Maya (Campos 1977), 22: Banos de Montemayor (Cäceres), 23: Herväs, (Cäceres), 24: Villasrubias (Salamanca), 25: Acebo (Cäceres), 26: Sabugal (MADUREIRA and MAGALHAESs 1980), 27: Belmonte (MADUREIRA and RAMALINHO 1981), 28. Miranda do Corvo (MADUREIRA and RAMALINHO 1981), 29: Caldas da Rainha (MADUREIRA and RAMALINHO 1981), 30: Obidos (RAMALINHO ın press), 31: Serra d’El Rei (RAMALINHO in press), 32: Rio Maıor (NIETHAMMER 1970), 33: Lourinhä (RAMALINHO in press), 34: Epinera-Cercal (NIETHAMMER 1970), 35: Tapada de Mafra (MADUREIRA and RAMALINHO 1981), 36: Torredeira (RAMALINHO in press), 37: Grijo (RAMALINHO in press), 38: Serra da Nogueira (RAMALINHO in press), 39: Cernadilla (HAusser 1984), 40: Cabeceiras de Basto (MADUREIRA and MAGALHAES 1980), 41: Geres (MADUREIRA and RAMALINHO 1981), 42: Lago de la Bana (Leön), 43: Panjön-Nigrän (Pontevedra), 44: Castrocaldelas (LorEz-Fuster 1983), 45: Caldas de Reyes (Pontevedra), 46: Puebla del Caramınal (LoPEZ-FusTER 1983), 47: Brandomil (LoPEz-FusTEr 1983), 48: Montesalgueiro (NoRESs 1979), 49: Betanzos (NORES 1979), 50: Vilaboa (CABRERA 1914), 51: Pontedeume (LoPEzZ-FUSTER 1983), 52: Caaveiro (NORES 1979), 53: Gontän-Quende (Lugo), 54: Montfero (LOPEZ-FUSTER 1983), 55: Ferrol (NOREs 1979), 56: Lagoa (Nores 1979), 57: San Cipriän (Lugo), 58: Estaca de Bares (Nores 1979), 59: San Esteban del Valle (Avila), 60: Monasterio de Yuste (Cäceres), 61: Estebanvela (ArenAs 1983), 62: Villacorta (Arenas 1983), 63: Pto. de la Quesera (ArEnas 1983), 64: Rıaza (ArEnAS 1983), 65: Sequera de Fresno (Arenas 1983), 66: Encinas (ArEnAas 1983), 67: Valdevacas (ArEnas 1983), 68: Sotillo (Arenas 1983), 69: Gallegos (Arenas 1983), 70: da Cuesta (ArEnas 1983), 71: Sotosalbos (ARENAS 1983), 72: Penalara (Arenas 1983), 73: Revenga (ArEnas 1983), 74: El Espinar (ArEnas 1983) 272 Table 2. Values of descriptive statistics of cranial variables in Sorex granarius from Galicia J. Gisbert, M. J. Löpez-Fuster, Rosa Garcia-Perea and J. Ventura Table 3. Values of descriptive statistics of jaw measurements of Sorex granarius from Galicia (GA), Gredos (GR) and Guadarrama (GU) a oo 0 a u N a u ON ON ON PN MNN NO SO oO Meam m N Mı-M; LM; EN 6.9 1.0 1.0 1.0 37 ar 3e 1.0 1.0 1.0 91 91 91 AG 42 43 Distribution and biometry of Sorex granarius 273 Table 4. Individual comparison between paired samples for each analysed variable, according to Tukey’s method <0.01 HC Results of Student-Newman-Keuls multiple test of rank for Galicia (GA), Gredos (GR) and Guadarrama (GU). 0 = Not significant differences cated by Mıtrer (1912), Hausser et al. (1975), MADUREIRA and MAGALHAES (1980), Nores (1979) and RAMALINHO (1981). Table 3 shows the statistics for the descriptive mandibular parameters of the Gadlıcıa, Gredos and Guadarrama samples. Comparison of the sample means (ANOVA test) discloses notable biometric sımilarıty among populations. Significant differences were detected only in & (F = 6.24, p < 0.01), LIA (F = 8.71, p < 0.01), SDI (F = 10.30, p < 0.01) and M,-M; (F = 4.76, p < 0.01). Individual comparisons between pairs of samples (Table 4) also confırm the scant divergence of peninsular Sorex granarıns populations. Between the populations of Galicia 274 J. Gisbert, M. J. Löpez-Fuster, Rosa Garcia-Perea and J. Ventura and Gredos, we only found statistically significant differences in LIA (p < 0.01), SDI (p < 0.01) and M,-M; (p < 0.01). Between the Galıcia and Guadarrama samples, the most marked divergences were in « (p < 0.05), LIA (p < 0.01) and SDI (p < 0.01). Between Gredos and Guadarrama, there was an even greater degree of biometric similarıty, only « showing significant differences (p < 0.01). In view of the mean values obtained for all these populations (see Table 3), the specimens from Galicia generally evidence the largest mandibular dimensions, although these differences lack statistical significance (observe values of ß, LIA, LM, SDI, LFT, M,-M;, LC and HC). Due to their biometrical similarity, the relationship between the sızes of the Gredos and Guadarrama samples has not been completely defined. To determine the degree of intersample affınıty ın each parameter, mean values were processed by the Student-Newman-Keuls test. In the graphic representation of results (Table 4), the populations with mean values not significantly heterogeneous are underlined (SoraL and RoHLr 1979). The resulting diagrams evidence the close relationship among the populations analysed. The only variable to statistically separate the Guadarrama population was &, which was smaller in this sample. The specimens from Galicia differ from Gredos and Guadarrama in LIA, SDI and M,-M;, the mean values of these parameters being higher in the Galicia samples. On the basıs of these findings, although there was no evidence of intraspecific mandibular biometric differences in Sorex granarius, there was a general tendency to smaller size from north to south. It is interesting to point out that in the Iberian Peninsula, Crocidura russula and Crocidura suaveolens have been found to present a clinal variation consisting in progressive reduction of cramıometric values to the south (Rey and LAanDın 1973; Rey and Rey 1974; Sans-CoMA et al. 1987). Acknowledgements We are grateful to M. NAvaRRoO and Dr. V. PEREZ-MELLADO, who contributed 7 specimens and six new localities, as to Dr B. THomas, who translated this manuscript to English, and I. MAIER for translating to German. Zusammenfassung Verbreitung und Biometrie von Sorex granarıus (Miller, 1910) (Soricinae: Insectivora) Die Verbreitung von Sorex granarins ist auf die Pyrenäen-Halbinsel beschränkt. Hier ist diese Art in Gallizien, der nördlichen Hälfte von Portugal und im Zentralgebirge anzutreffen. Sorex granarıns bewohnt vorwiegend Waldgebiete. Die vertikale Verbreitung erstreckt sich von der Meereshöhe bis zur Höhe von 2000 m. Im Zentralgebirge ist dieses Intervall kleiner (500-2000 m). Die Biometrie der Schädel von untersuchten Individuen ist sehr gleichartig, obwohl sich eine Tendenz zur Größenab- nahme von Norden nach Süden beobachten läßt. Die größten Schädel wurden in Gallizien gefunden. Literature Armaca, C. (1968): La faune mammalogique du Portugal dans la checklist of Paleartic and Indian Mammals, 1951, par Ellerman et Morrison-Scott. Arg. Mus. Boc., 2 serie, 12. ARENAs, A. (1983): Atlas de Micromamiferos de Segovia. Un modelo de anälisis zoogeogräfico de Insectivoros y Roedores mediante egagröpilas de lechuza. Tesina, Univ. Madrid. CABRERA, A. (1914): Fauna Iberica. Mamiferos. Mus. Nac. Ciencias Nat.: Madrid. Campos, F. (1977): Regimen alimenticio de Tyto alba en las provincias de Salamanca y Zamora. Ardeola 24, 105-119. CATZEFLIS, F. (1984): Systematique biochimique. Taxonomie et phylogenie des musaraignes d’Europe (Soricıdae, Mammalıa). These, Universit€ Lausanne. CATZEFLIS, F.; GRAF, J.-D.; HaussEr, J.; VOGEL, P. (1982): Comparaison biochimique des Musaraig- nes du genre Sorex en Europe occidentale (Soricidae, Mammalia). Z. zool. Syst. Evolut.-forsch. 20, 223-233. Dixon, W.J. (1983): BMDP Statistical Software. Berkeley: University of Calıfornia Press. Distribution and biometry of Sorex granarius 272 GARZON-HEYDT, J.; Castroviejo, $.; Castroviejo, J. (1971): Notas preliminares sobre la distribuciön de algunos micromamiferos en el norte de Espana. Säugetierkdl. Mitt. 19, 217-222. Hausser, J. (1976): Contribution A l’etude des Musaraignes du genre Sorex (Cytotaxonomie, Morphologie, Repartition). These, Universite Lausanne. — (1984): Genetic drift and selection: Their respective weigths in the morphological and genetic differentiation of four species of shrews in Southern Europe (Insectivora, Soricidae). Z. zool. Syst. Evolut.-forsch. 22, 302-320. HAUSSER, ]J.; CATZEFLIS, F.; MEYLAN, A.; VOGEL, P. (1985): Speciation in the Sorex araneus complex (Mammalia: Insectivora). Acta Zool. Fennica 170, 125-130. HAUSSER, J.; GRAF, J.-D.; MeyLan, A. (1975): Donnees nouvelles sur les Sorex d’Espagne et des Pyrenees (Mammalıa, Insectivora). Bull. Soc. Vaud. Sc. Nat. 348, 241-252. HausseER, ]J.; JAMMOT, D. (1974): Etude biometrique des machoires chez les Sorex du groupe araneus en Europe continentale (Mammalia, Insectivora). Bull. Soc. Vaud. Sc. Nat. 348, 241-252. HEım DE Barsac, H.; DE BEAUFORT, F. (1969): Contribution A l’etude des micromammiferes du Nord-ouest de l’Espagne (Santander, Asturies, Galice, Leon). Mammalia 33, 367-380. JammoT, D. (1973): Mise au point d’un “bloc comparateur optique’” pour les neccessites d’une osteometrie fine. Mammalıa 37, 122-125. LoPEZ-FUSTER, M. J. (1983): Sobre los generos Sorex Linnaeus, 1758, Suncus Ehrenberg, 1833 y Crocidura Wagler, 1832 (Insectivora, Soricidae) en el Nordeste de la Peninsula Iberica. Tesıs Doctoral, Universidad Barcelona. MADUREIRA, M. L.; MAGALHAESs, C. M. (1980): Small mammals of Portugal. Arq. Mus. Bocage, 2 ser., 7, 179-214. MADUREIRA, M. L.; RAMALINHO, M. G. (1981): Notas sobre a distribuicao, diagnose e ecologia dos Insectivora e Rodentia portugueses. Arg. Mus. Bocage, ser. A, 1,165-263. MALEc, F.; STORCH, G. (1964): Einige Kleinsäuger (Mammalia: Insectivora, Rodentia) aus Nordspa- nıen. Z. Säugetierkunde 29, 220-230. MEyLAnN, A.; Hausser, ]J. (1973): Les chromosomes des Sorex du groupe araneus-arcticus (Mammalıa, Insectivora). Z. Säugetierkunde 38, 143-158. MILLER, G. $. (1910): Description of sıx new European Mammals. Ann. Mag. Nat. Hist., Ser. 8, 6, 458461. — (1912): Catalogue of the Mammals of Western Europe. London: British Museum (Nat. Hist.). NIETHAMMER, J. (1956): Insektenfresser und Nager Spaniens. Bonn. Zool. Beitr. 7, 249-295. — (1970): Über Kleinsäuger aus Portugal. Bonn. Zool. Beitr. 21, 89-118. Nores, C. (1979): Nuevas aportaciones al conocimiento de la subfamilia Soricinae (Mammalıa, Insectivora) en los distritos Cantabrico y Lusitano. Tesıs Licenciatura, Universidad Oviedo. RAMALINHO, M. G. (1981): Portuguese Soricinae. A review of recent data (in press). Rey, J. M.; Lanpın, A. (1973): Sobre la presencia de Crocidura suaveolens en el sur de Andalucia (Mammalıa, Insectivora). Bol. R. Soc. Esp. Hist. Nat. (Biol.) 71, 9-16. Rev, J. C.; Rey, J. M. (1974): Nota preliminar sobre las musaranas del genero Crocidura Wagler, 1832 en las islas Baleares. Bol. Est. Centr. Ecol. 3, 79-85. Rıvas-MARTINEZ, S. (1981): Les etages bioclimatiques de la vegetation de la PEninsule Iberique. Anal. Jardın Botänico de Madrid 37, 251-268. — (1983): Pisos bioclimaticos de Espana. Lazaroa 5, 3343. Sans-Coma, V. (1979): Beitrag zur Kenntnis der Waldspitzmaus, Sorex aranenus Linne, 1758, ın Katalonien, Spanien. Säugetierkdl. Mitt. 27, 96-106. SANS-CoMA, V.; LOPEZ-FUSTER, M. J.; VarGas, J. M. (1983): Sobre la musarana comün, Crocidura russula (Hermann, 1780) en el sur de la Peninsula Iberica. In: Mamiferos y Helmintos. Ed. by M. Bacn. Barcelona (in press). SoRKAL, R. R.; RoHLF, F. J. (1979): Biometria. Barcelona: Ed. Blume. Authors’ addresses: J. GIsBERT and Rosa GARCIA-PEREA, Unidad de Zoologia Aplıcada, El Encin, Apdo. 127, Alcalä de Henares, Madrid, Spain; M. J. LöpEz-Fuster and ]. VENTURA, Departament de Vertebrats, Facultat de Biologia, Av. Diagonal 657, E-08028 Barcelona, Spain Z. Säugetierkunde 53 (1988) 276-280 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Notes on some bats from northern Thailand, with comments on the subgeneric status of Myotis altarıum B. R. BLoop and D. A. McFARLANE Section of Birds and Mammals, Los Angeles County Museum of Natural History, Los Angeles Receipt of Ms. 06. 04. 1987 Abstract Reported is the occurance of four species of bats from extreme northern Thailand collected by members of the National Speleological Society, USA, Thailand Karst Hydrologic Expedition. Single specimens of Aselhscus stoliczkanus, la io, Miniopterus macrodens, and Myotis altarınm were collected. This is the first record of M. altarınm outside of China and only the second record for 7. io. Although known from the area, our specimen of A. stoliczkanus does not match published color patterns of this species from Thailand. We provide the first published photograph of the skull of M. altarinm and an expanded description of this species because previously published descriptions are inadequate to place this species at the subgeneric level. We compared M. altarınm directly to members of three other subgenera of Myotis and conclude it belongs in the subgenus Myotıs. Introduction The Thailand Karst Hydrologic Survey, fielded by members of the National Speleological Society between December 1983 and February 1984, provided several specimens of cave- dwelling bats from the limestone mountains of extreme northern Thailand. We report here three species of Vespertilionidae plus one species of Hipposideridae. We found published descriptions of one of the species collected, Myotis altarınm, inadequate to properly place it at the subgeneric level. Therefore, we present an expanded description of M. altarınm, as well as, the first photographs of its skull. Specimens have been deposited in the mammal collection of the Natural History Museum of Los Angeles County (LACM). The rhinolophids collected during this expedition are reported separately (MCFARLANE and BLoop 1986). Results and discussion Systematic accounts Aselliscus stoliczkanus Dobson, 1871 (Hipposideridae) Specimen collected; LACM 70301; collected by R. HEMPERLY, 24 Dec. 1983; male, ın alcohol with skull extracted. Northern Thailand; Chang Dao Mountain, Chang Dao Cave. 19°23’N, 98°54’E. Some selected measurements (in mm) are head and body length, 43; taıl length, 38; forearm length, 41; greatest length of skull, 18; hind foot length, 5; ear length, 8. LEkAGUL and McNEELEY (1977) report this species to be uncommon, but wide spread. They further report that this species occurs in two color phases, a brown-gray phase and a yellow-red phase. The LACM specimen has long dorsal fur (7 mm) which is a creamy white at its base with rich dark brown tips. The ventral fur color is a silvery brown. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5305-0276 $ 02.50/0 Notes on some bats from northern Thailand UL Miniopterus macrodens Maeda, 1982 (Vespertilionidae) Specimen collected: LACM 70323; collected by R. HEMPERLyY 12 Jan. 1984; male, in alcohol with skull extracted. Northern Thailand, Aung Kang Region, Big House Cave; 19°50'N, 99°10’E. Selected measurements (in mm) are as follows; total length, 107; forearm length, 50.5; greatest length of skull, 16.4; ear length 16; tragus length, 4.5; length of third metarcarpal, 47.5; length of first phalanx of digit 3, 11.9; length of second phalanx of digit 3, 34.5. Our specimen is all black in color including the flight membranes. This species is part of the M. maginater group (R. PETERSON, pers. comm.) la io Thomas, 1902 (Vespertilionidae) Specimen collected: LACM 70321; collected by R. Hemperıy 11 Jan. 1984; in alcohol. Northern Thailand, Aung Kang Region, Big House Cave. 19°50’N, 99°10’E. Some representative measurements (in mm) include head and body length, 90.6; tail length, 71.5; forearm length, 75.3; ear length, 26; tragus length, 8.5; length of metacarpal III, 68.3; length of third phalanx of digit III, 11.9. This specimen was collected by hand near the entrance of Big House Cave. The field notes indicate that it was not in the immediate company of other bats. However, three other species were collected in this same cave: Rhinolophus malayanus, R. robinsoni, and M. altarınm (MCFArLANE and BLooD 1986). Only one other specimen (Museum of Comparative Zoology 3549) is known from northern Thailand, near Chang Dao (Arzen and CooLipGe 1940). Myotıs altarınm Thomas, 1911 (Vespertilionidae) Specimen collected: LACM 70234, collected by J. BEnepıcr 19 Jan. 1984; Northern Thailand, Aung Kang Region, Big House Cave; male, in alcohol with skull extracted. This single specimen was collected approximately one meter above a stream, on the cave wall. The field notes do not indicate whether or not this specimen was near other bats. This specimen represents the first record for this species outside of mainland China and extends the known range for this species by 1250 km southwest. The two other known localities are Omi San, Szechwan Province, China (the type localıty) and 3 mi. east of Kweiyang City in Kweichow Province, China (STAGER 1949). Two of STAGER’S six specimens (LACM 8214 and 8215) compare favorably with the new specimen. Selected measurements of LACM 70234, 8214, 8215, and the type (as given by THoMmas 1911) respectively are as follows: head and body length, 55, 60, 58, 55; torearm length, 44, 45.35, 42.90, 45; tibia, 16.50, 18.65, 18.10, 29 (reported as lower leg plus foot), hindfoot length, 11, 12, 12, none repoted for type, ear length, 22, 24, 24, 22; greatest length of skull, 15.50, 15.95, 16, 15.2. In addition to M. altarınm, seven other species of Myotis are known from northern Thailand (LexaGurL and McNEELY 1977): M. mystacinus, M. annectans, M. yosseti, M. siligorensis, M. horsfieldü, M. chinensis, and M. hasseltüi. Published information on the relationships of M. altarınm are contradictory because so little material is available. THomas (1911) stated that the morphology of M. altarınm most resembled M. pequinus, a member of the large footed subgenus, Lenuconoe. TATE (1941) placed M. altarınm in the M. emarginatus section of the subgenus Selysius. The other species in this section of Selysins (sensu TATE 1941) are M. emarginatus, M. peytoni (= montivagus; Honackli et al. 1982), M. primula (= annectans; Honackl et al. 1982), and M. saturatus (= emarginatus; CORBET 1978). FinpLey (1972) also placed M. altarıum ın the subgenus Selysins, but this analysıs placed M. altarınm close to a different group of species: M. mystacinus, M. sıligorensis, M. davidır (= mystacınus; Honackt et al. 1982), and M. ikonnikovi, FINDLEY’sS (1972) analysıs and later classification placed the species M. emar- ginatus and M. pequinus into the subgenus Myotis. The relationships of M. altarınm are 278 B. R. Blood and D. A. McFarlane therefore confused, because according to TATE (1941) it should be placed close to M. emarginatus and so should be ın the subgenus Myotis as this subgenus is characterized by FInDLEy (1972). Based upon the three LACM specimens and THomas (1911), M. altarium can be characterized as follows: a medıum sized Myotis with ears twice as long as they are wide, and which are distinctively black and translucent. The lateral edge of the ears are folded cranıally. When laıid forward the ears reach 5 mm beyond the tip of the muzzle. There is a distinct lobule at the base of the ear. The lobule is separated from the ear by a deep notch. The tragus is long, thin, and bluntly pointed. The flight membranes are naked, blackish brown ın color and attach to the hind foot at the distal end of the metatarsal of digit one. There are eight striae on the uropatagium, which ıs naked dorsally, but with scattered hairs ventrally. The dorsal fur is light brownish ın color and 12-15 mm ın length. The ventral fur color is lighter having dark bases and whitish tips. The calcar is weakly lobed at its mid point. The hind foot measures over 60% ot tibial length. The skull of Myotis altarınm (Fig. 1) has a short and distinctly upturned rostrum in lateral view. A line drawn along the long axis of the rostrum intersects a line drawn along the long axis of the cranıum at approximately 145°. The skull slopes sharply anteriorly to a point above the posterior premolars. The skull profile has a slıght convexity at mid- braincase (where the frontal and parıetal bones meet). The nasal bones are flattened proximally and slightly arched (lateral to medial) distally and curved upwards distally at the premaxillary-nasal junction. The junction ot the premaxillary-nasal bones is at the level ot the superior surface of the orbit. The posterior part of the premaxillary bones are arched and display a obvious bump in lateral view. The zygomatic arches are consistant ın depth, as seen in lateral view and expand somewhat posteriorly in dorsal view. The anteorbital foramen opens 0.7 mm anterior to the orbit above the anterior labial root of P° and is 0.3 mm ın diameter. The anteorbital foramen leads to an enclosed basın found above the roots of the premolars. Fig. 1. Dorsal, lateral, and ventral views of the cranıum of M. altarınm. (Approximately 3x lıfe size) The ventral skull surface displays several distinctive features. The palate is deeply vaulted. An unusual feature of the palate is a sharp ridge running transversely between the canine teeth. The foramen ovale is large and conspicuous in ventral view. The basısphenoid is raised above the floor of the interbullar area resembling a bridge. There is no sagıttal crest on any of the specimens. All upper and lower premolars are fully in line with the Notes on some bats from northern Thailand 279 toothrow. The dental formula is 2/3-1/1-3/3-3/3. The ratio of the length to width of the M?° is 1:3. The crown of the M? is characterized by not possessing a ridge connecting the hypocone and metacone. Because of the orientation of the rostrum the incisors and canines are directed obliquely anteriorly. We have examined both TATE (1941) and FinpLey (1972) ın compiling a provisional list of characteristics of these three subgenera of Myotis. Although it ıs not within the scope of this paper to attempt to fully characterize all the subgenera of the genus Myotis we brıefly list some of the most important characteristics of the subgenera needed to evaluate M. altarıum. M. altarınm is not a member of the large-footed subgenus Leuconoe because it does not possess a protoconule on its upper molars, a key characteristic of this subgenus (TATE 1941). Species within the subgenus Sel/ysius possess at least the following: small feet (< 50 per cent ot the tıbıa length), wing membrane to base ot fırst toe, calcar with lobe, ear small, low rostrum with abruptly rising frontal region, length to width ratio of M° is 0.8 to 1.5, small overall size, large uropatagial traps, uropatagıal hairs restricted to striae, anteorbital foramen diameter near 0.6 mm. The species included within the subgenus Myotis possess at least the following: large species with the feet not enlarged (hind foot at least 60 per cent of tibia length), ears can be elongate, broadened or both, rostrum is low with a gradually rising frontal region, uropatagium nearly naked with hairs scattered widely over ıts surface not restricted to striae, length to width ratio of M’ is 1 to 2.7, anteorbital foramen is close to orbit. In order to best compare the above characteristics to M. altarınm we examined members of each subgenus and compared them directly to M. altarium (see specimens examined). The dorsal skull profile and the longıtudinal sulcus between the nasal bones ın M. altarınm are intermediate between M. myotıs and M. muricola. This sulcus is deep ın M. emarginatus and M. murıicola and essentially missing in M. myotis. The structure of the zygomatic arches, crown pattern of M?, disposition of the anteorbital foreamen, distribu- tion of uropatagial hair, attachment of the wing to the foot, length to width ratio of M’, relative length of its hind foot, and ear structure most closely resemble the condition of the subgenus Myotis. In conclusion, based upon the comparisons we have made M. altarınm ıs best considered closest to the subgenus Myotis, but our comparison has been too narrow to place M. altarınm close to any one species of Myotıs. Specimens examıned Myotıis altarınm (3); LACM 70234, 8214, 8215, localıties in text: Myotıs emarginatus (5); LACM 58397-58401, Portugal: Myotis thysanodes (1); LACM 55951, Colima, Mexico: Myotis daubentoni (1); LACM 58781, Germany: Myotis muricola (6); American Museum of Natural History (AMNH) 102967-102972, Sumatra: Myotıs myotis (6); AMNH 150102-150107, Bavarıa, Germany. Acknowledgements We would like to thank the members of the Karst Hydrologic survey team, especially B. BENEDICT and R. HEMPERLEY for collecting and donating the specimens upon which this report is based. D. PATTEN, J. MATSON, L. BARKELY, and $. GEORGE read early versions of the manuscript and provided helpful comments and discussion. We wish to thank K. Koopman and M. Bocan for thoughtfully reviewing the manuscript. We are very grateful to R. PETERSON who identified the specimen of Miniopterus. Many thanks also go to C. BLoop for reading and proofing the many versions of this work. The photographic work was performed by ]J. DELEon and D. MEIER and the figure was prepared by C. Maroor of the Natural History Museum of Los Angeles County. We are grateful to I. KRATTIGER who translated the summary. 280 B. R. Blood and D. A. McFarlane Zusammenfassung Über einige Fledermäuse aus dem nördlichen Thailand mit Bemerkungen zur subgenerischen Zugehörigkeit von Myotis altarınm Über folgende, im äußersten Norden Thailands Ende 1983 bis Anfang 1984 gesammelten Fleder- mausarten wird berichtet: Aselliscus stoliczkanus, la io, Miniopterus macrodens und Myotis altarium. Unser Exemplar von Myotis altarınm bildet den ersten Nachweis dieser Art außerhalb Chinas. Wie eine ausführliche Beschreibung ergibt, dürfte M. altarınm ın die Untergattung Myotis der Gattung Myotis gehören. Literature Arten, G. M.; CooLiDGe, H. $. (1940): Mammal collections of the Asiatic Primate Expeditions. Bull. Mus. Comp. Zool. 97, 131-166. CoRBET, G. B. (1978): The mammals of the Palaearctic region: a taxonomic review. New York: Cornell Univ. Press. FinDLe£v, J. S. (1972): Phenetic relationships among bats of the genus Myotis. Syst. Zool. 21, 31-52. Honackı, J. H.; Kınman, K.E. G.; Koeppr, J. W. (1982): Mammal species of the world. Lawrence: Allen Press. LEKAGUL, B.; McNEELYy, J. (1977): Mammals of Thailand. Bangkok: Kurusapha Ladprau Press. MCFARLANE, D. A.; Broop, B. R. (1986): Taxonomic notes on a collection of Rhinolophidae (Chiroptera) from northern Thailand, with a description of a new subspecies of Rhinolophus robinsoni. Z. Säugetierkunde 51, 218-223. STAGER, K. E. (1949): Notes on the mammals of Kweichow Province, China. J. Mammalogy 30, 68-71. TATE, G. H. H. (1941): Results of the Archbold Expeditions. No. 39. Review of Myotis of Eurasia. Bull. Amer. Mus. Nat. Hist. 78, 537-565. THomas, O. (1911): The Duke of Bedford’s zoological exploration of Eastern Asia. XIII. On mammals from the provinces of Kan-su and Sze-chwan, Western China. Proc. Zool. Soc. London, 158-180. Authors’ address: Brap R. BLooD and DonALD A. MCFARLANE, Section of Birds and Mammals, Los Angeles County Museum of Natural History, 900 Exposition Blvd., Los Angeles, California 90007, USA Z. Säugetierkunde 53 (1988) 281-293 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Search calls of echolocating Nyctalus leisleri and Pipistrellus savii (Mammalia: Chiroptera) recorded in Switzerland By P. E. Zınas Institute of Zoology, University of Berne Receipt of Ms. 06. 04. 1987 Abstract Studied the search calls of hunting Nyctalus leisleri and Pıpistrellus savıi north and south to the Swiss Alps. The ultrasounds were converted to audio frequencies by a 8:1 dividing detector and subsequently analysed in the laboratory on a dual channel digital oscilloscope using a period meter. In both species single harmonic pulses with shallow frequency modulation and pulses of constant frequency were recorded. Search pulses of N. leisleri (median values of 204 signals): pulse duration 9.2 ms; start frequency 25.1 kHz; center frequency 24.3 kHz; end frequency 23.7 kHz; sweep bandwidth 1.6 kHz. The intervals (N = 2153) between search pulses followed a bimodal distribution with a higher mode at 243 ms and a weaker one at 350 ms. Search pulses of P. savır (median values of 236 signals): pulse duration 12.1 ms; start frequency 34.2 kHz; center frequency 32.8 kHz; end frequency 32.0 kHz; sweep bandwidth 2.2 kHz. The intervals (N = 1341) between search pulses followed a bimodal distribution with a higher mode at 192 ms and a weaker one at 294 ms. In both species various signal varıables were significantly correlated. In addition to echolocation signals two types of socıal calls of N. leisleri were recorded. The first type consisted of a group of 4 pulses ranging from 38-16 kHz. The second type was a single, linear period modulated pulse of 20-27 ms duration, ranging from 21-10 kHz. Introduction In the course of a study about the acoustic ıdentification of certain bat species ın Switzerland, tape recordings of the acoustic behaviour of Nyctalus leisleri and Pipistrellus savıı were made. Until now only little information on the echolocation sounds of both species had been available. HooPEr (1969) described the approximate frequency range and pulse repetition rate of captive N. leislerı. KonsTAnTINoV and MARAROV (1973), and PATLJAKEVITSCH (1980) studied echolocation pulses of captıve P. savıi. The present paper describes search calls of free flying individuals of both species as well as cries of N. leislerı which are assumed to be socıal calls. Material and methods Search pulses of free flying N. leisleri were recorded in the southern Alps (N = 38 analysed sıgnals) and northern alps (N = 140) of Switzerland. One N. leisleri, marked with reflective tape on a ring, was recognized some weeks after release in the northern Alps. Search pulses were recorded, when thıs bat was hunting above a road with street lights. A further batch ot signals (N = 26) ot a single individual in the plane near Berne has been analysed. For statistical analysis all the recordings were pooled (N = 204 analysed signals). Search pulses of free flyıng P. saviı were recorded in northern (N = 73 analysed signals), central (N = 90) and southern (N = 73) parts of the Swiss Alps. For statistical analysıs all the recordings were again pooled (N = 236 analysed sıgnals). “The distinction between search and approach calls is arbitrary as they are part of a con- tinuum. ..” (FEnTon and Ber 1981). For thıs study only pulses being part of more or less regular U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5305-0281 $ 02.50/0 282 PER ZinER sequences were analysed. Pulses whose BW/T-factor (kHz/ms) exceeded a value of 1.0 were not considered as search pulses and excluded from calculations. In addition to the above signal sample obtained from free flying bats, echolocation pulses of several individuals of P. saviz and N. leisleri were collected, when the bats were released after net trappıng. The aım was to get reference sıgnals from known individuals in order to facilitate the acoustic recognition of free flying bats. All ultrasounds were picked up with a modified QMC S100 ultrasound detector connected to a frequency divider, developped and built by K. ZBInDEn, University of Berne. The frequency divider produces a sinusoıdal waveform at one eighth of the original frequency and retains the envelope of the original signal. Further specifications are: Frequency range 10-220 kHz (+/-2 dB), S/N-ratıo 55 dB (lin. 10-220 kHz), switchable HP-filter (20 kHz, 24 dB/octave). The transformed ultrasounds were recorded with aSONY TC-D 5M on metal cassettes (-3 dB at 15 kHz). The recordings were analysed on a Nicolet type 3091 dual channel digital oscilloscope (2x4K words) using a period meter (zero crossing detector, developped by D. HArTLEY, Queen Mary College, London and K. ZBInDEN, University of Berne, who also built the instrument). Of each pulse the following signal variables were measured: The start frequency (fs), the instantaneous frequency at maximum signal amplitude (fmax), the center frequency (fc) at half the pulse duration and the end frequency (fe). The frequency varıables were estimated to an accuracy of +/-0.4 KHz in realtime. The sweep bandwidth was calculated as BW = fs-fe. The pulse durations were estimated to an accuracy of approx. 0.2 ms and intervals between pulses (from pulse center to pulse center) to an accuracy of +/ =2.ms. Averaged power spectra were obtained on an FFT-analyser type 2033 of B& K, Denmark (400 lines, flat weighting, linear average mode, RMS spectrum). Computations of quartiles and modes were made according to SacHs (1978). Preceeding the computations all the measurements were classified using class widths of 10 ms for interval durations, 1 ms for pulse durations and 0.8 kHz for frequencies (real time). Goodness-of-fit tests for normal distribution, correlation coefficients and arıthmetic means were computed on a HP-71B calculator, using ıts AMPI stastistics library. In those cases where the distribution did not conform to the standard probability models, the mediıan instead of the arıthmetic mean was taken as a representative measure of location (or of central tendency). The interquartile-range Q.o.757-Q.o.25), which contains 50% of a sample, was used as a measure of dispersion. To allow a comparison with the data obtained by other authors, both the parametric and nonparamerric statistics were calculated. Results Nyctalus leisleri Intervals between search pulses showed a bimodal distribution and varıed considerably (Fig. 1A). A first prominent mode was at 243 ms and a second weaker mode at 350 ms (both modes were estimated using the original class width of 10 ms). The pulse durations were normally distributed (x’=9.5, df=10) with a median of 9.2 ms (Fig. 1B). The distributions of the four frequency varıables are shown in Figure 2, their statistical parameters in Table 1. A goodness-of-fit test was consistent with normal distribution in the case of the center frequency (X=12.3, df=9), the maximum frequency (x’=12.2, df=8) and the end frequency (x’=2.8, df=7), but not the bandwidth (x’=351, df=8) and the start frequency (X°=71.3, df=14), which skewed towards the higher frequencies. In this sample the end frequency had the smallest statistical dispersion. The averaged power spectrum of 50 search pulses showed a dominant frequency at 23.8 kHz and a-6 dB spectrum bandwidth ot 2.9 kHz (Fig. 3). Most search pulses of N. leisleri were frequency modulated by a small amount only. Nine of 204 pulses (T=3.7-12.2 ms) were of constant frequency (CF) in the range of 22.4-27.2 kHz. In the distribution of bandwidth (BW) the modal class and both classes on each side of it (range 0.8 to 2.4 kHz) contained 76% ot the analysed pulses (see Fig. 1C). Pulses emitted during the approach phase however, frequently had a very steep frequency modulation and a bandwidth of more than 30 kHz. Search pulses of N. leisleri often increased rapidly to a high amplitude level at their beginning (Fig. 4). When listening to the bats by means of a broadband detector this 20 DJ H<100%2) = 2153 Mode 1 = 243 ms 23 4% 33% B c 30 20 - 25 NC100%) = 204 7 H<100%>3 = 204 15 Median = 9.2ms 20 Median = 1.6 kHz 10 pulse duration T ms Bandwidth (BH) kHz Fıg. 1. Nyctalns leisleri. A: Interval duration between search pulses (from pulse center to pulse center). Width ot classes is 20 ms. B: Duration of search pulses. Width of classes is 1 ms. C: Sweep bandwidth (BW) of search pulses. Width of classes is 0.8 kHz 30-% il. A B 25 25 H(100#) = 204 If N<1007>3 = 204 20 20 Median = 25.1 kHz An Median = 24.6 kHz start frequency (fs) kHz max. frequency (fmax) kHz 30 % c 30007 In D 25 H{1004> = 204 sei) Median = 24.3 kHz 20 20 15 15| N<1007) = 204 10 10 | Median = 23.7kHz I k | 0 E 0 18 20 24 28 32 34 18 20 24 28 See 3d center frequency (fc) kHz end frequency (fe) kHz Fig. 2. Frequency variables of search pulses of Nyctalus leisleri. Width of classes is 0.8 kHz. A: Start frequency at the beginning of a pulse. B: Instantaneous frequency at maximum signal amplitude. C: Center frequency at half the pulse duration. D: End frequency 284 PIE ZIngE; Table 1. Parameter values of search pulses (N = 204) of Nyctalus leisleri, as shown in Figs. 1 and 2 Median Q3-Q1 95 % confidence interval of x Pulse duration T (ms) 3.3 oz Start frequency fs (kHz) 2.9 25.3-25.9 Maximum frequency fmax (kHz) 2.0 24.5-24.9 Center frequency fc (kHz) 23.4 j 5 29 24.2-24.7 End frequency fe (kHz) 22.8 287 1.8 23.5-23.9 Bandwidth BW (kHz) 1.0 1.6 2,5 1.5 L Qi and Q3 are the quartiles Qo.5, and Q.o.5, of the distributions. Q3-Q1 is the interval that contains 50 % of the whole sample (or 50 % of the frequency distribution area). Fıg. 3. Nyctalus leisleri. Averaged power spectrum of 50 search pulses (400 lines, RMS-spectrum type, flat weighting, linear averagıng). Frequency span: 8-48 kHz. The vertical cursor is at the dominant frequency of 23.8 kHz. The -6 dB spectrum bandwidth ıs 2.9 kHz resulted in an explosive sound, a phenomenon not found to the same extent ın other bat species of Switzerland. Due to the recording system used, the presence of harmonics could not be checked systematically. In a few search pulses ‘steps’ in the period plot seemed to indicate the eventual presence of a strong second harmonic, however. It is known that there may exist relationships e.g. between pulse duration and bandwidth, depending on the function of a pulse (TuPpınIEr et al. 1980; HARTLEY 1985). Correlation coefficients were computed (Tab. 2) to check interesting relations between the measured varıables of search pulses. Negative correlations between pulse duration T and the four frequency varıables were found. This indicates that shorter search pulses tended to have higher values in all the frequency varıables. There was no significant correlation between bandwidth and search pulse duration. This and the positive correlation between the end and the start frequency means that short pulses as a whole were situated higher in the frequency scale. Search calls of echolocating Nyctalus leisleri and Pipistrellus savıı 285 Fig. 4. Characteristic search pulse of Nyctalus leisleri. Oscillogram with superimposed fre- quency plot. I=8.2 ms, fs=23.2 kHz, fmax=22.4 kHz, fc=22.4 kHz, fe=22.4 kHz. The vertical cursor indicates the pulse center (half duration). The horizontal cursor is at 30 kHz in the frequency plot Table 2. Correlation coefficients (arranged from highest to lowest) of selected variables of a random subsample of 49 search pulses of Nyctalus leisleri ntests)) 20.777. r(de, 1) = 0.653 r(BW, T) = +0.001 Rates), = 07239 r(fs, 1) = -0.554 e(imaz, TE) = 0.667 r(BW, fe) = +0.037 A goodness- -of-fit test was consistent with normal ‚distribution for all the variables. T: ?=8.8 (df=8). fs: x’=2.5 (df=5). fmax: y’=2.4 (df=6). fc: y’=3.8 (df=4). fe: y’=3.8 (df=5). BW: x? =7.4 (df=3). The null hypothesis, Ho: p=0, can be rejected at P<0.001, for all the r-values, except r (BW, fe) and r (BW, T), where P>0.1 Social calls of N. leisleri On 4 July 1986 at 2:30 MEZ, two N. leisleri were hunting insects above the street lights at Meiringen (Northern Alps). In addition to echolocation signals, complex patterns of cries, probably of social function, were recorded. It is not known whether the calls were emitted by both animals. Two types of social calls will subsequently be described. The first type of social call consisted of a group of four pulses (Fig. 5). Four of these calls were emitted in short sequence separated by intervals of 195-485 ms, containing a number of echolocation pulses. u Fıg. 5. Nyctalns leisleri. Social call type 1. Oscillogram with superimposed frequency plots (horizontal axıs 4.0 ms/div). Cf. text for further details 286 P. E. Zıngg With about 9 ms the first cry of a group had the longest duration. It used to start at 38 kHz with a short, shallow frequency modulation, followed by a long, steeper part and ended at 22 kHz with shallow modulation again. In one group the change in the modulation rate of the first pulse occurred abruptly and resulted in an “elbow”-shaped frequency plot (Fig. 5). The following three pulses in the groups were all different from the first one. Their pulse duration was shorter and varied from 1.6 to 4.8 ms. In each group the last pulse was the shortest. The second, third and fourth pulse started at 20-27 kHz with a maximum amplitude at approx. 21 kHz and ended at 16-20 kHz. They, and especially the fourth pulse in each group, appeared to include higher harmonics of varyıng strength. The intervals between pulses were in the range of 4.7-11.5 ms. In late August 1986 similar socıal calls were recorded in southern Switzerland. They were emitted by two or three unknown bats which stayed in a group of chestnuttrees (Castanea sativa). The calls consisted of three to fıve pulses. The first pulses had a duration ot 9-14 ms. They started with shallow frequency modulation at 25-28 kHz. The modula- tion rate became steeper ın the second half of the pulse which ended at a frequency of 18-20 kHz. The plot of the instantaneous frequency showed a convex curve. The following pulses in a group always had a shorter duration (1-5 ms). Their lowest fre- quency varıed from 14 to 21 kHz. At Meiringen a second type of social call of N. leisleri was recorded three times, when a pipistrelle bat (P. pipr- strellus) passed (Fig. 6). The pulse dura- tion varıed from 24 to 27 ms. All pulses were more or less linear period mod- ulated, started at 18-17 kHz and ended at 11-10 kHz. The FFT power spectrum of the pulse shown in Figure 6 had a peak at 12.6 kHz and a -6 dB spectrum : I AVDARE | bandwidth of 1.4 kHz. Fig. 6. Nyctalus leisleri. Social call type 2. Oscillo- 2 The same type of social call was re- gram with superimposed frequency plot (horizontal axis 5.4 ms/div.). T=24 ms, fs=17 kHz, fmax=fc=13 corded in southern Switzerland from kHz, fe=10 kHz the same two or three unidentified bats as described above. Mean values +/- standard deviation of 22 analysed sıgnals from southern Switzerland were: T=22.1 +/-2.17 ms, fs=19.2 +/-0.78 kHz, fmax=12.5 +/-0.57 kHz, fc=13.7 +/ -0.57 kHz, fe=11.2 +/-0.39 kHz and BW=8.0 +/-0.63 kHz. These parameter values are close to those of the calls recorded in northern Switzerland. Pipistrellus savıi The distribution ot intervals between search pulses (Fig. 7A) was again multimodal and had a prominent peak at about 190 ms. There was a second much smaller mode at 294 ms (both modes were computed using the original class wıdth of 10 ms). For all the varıables T, fs, fmax, fc, fe and BW the hypothesis of normal distribution could not be maintained (Chi-square goodness-of-fit test). Thus the median and the quartiles are better suited to characterize the distributions. In the distribution of pulse durations (median: 12.1 ms) 50% of the sample were between 10.9 ms (Qo 25) and 13.8 ms (Qo.5) (Fig. 7B and Tab. 3). Two pulses in the sample were considerably longer (20 and 21 ms). Although their BW/T is not higher than 1.0, five pulses of the sample may be classıfied Search calls of echolocating Nyctalus leisleri and Pipistrellus savü 287 N{£100%5 = 1341 Mode 1 = 192 ms Mode 2 = 294 ms 200 00 ms Interyal duration between search pulses NC100%) = 236 Median = 12.1ims 20 N{100%) = 236 Median = 2.2 kHz Di 15 10 be) hr] 0 0 pulse duration T MS Bandwidth (BW) kHz Fıg. 7. Pipistrellus savü. A: Interval duration between search pulses (from pulse center to pulse center). Width of classes is 20 ms. B: Duration of search pulses. Width of classes ıs 1 ms. C: Sweep bandwidth (BW) of search pulses. Width of classes is 0.8 kHz as approach pulses when their high start frequencies and bandwidths are considered (Figs. 8A and 7C). Parameter estimation excluding these five outliers does not shift the obtained results substantially, however. The end frequency had a very prominent peak at 32 kHz and the smallest dispersion (Fig. 8D). The center frequency (Fig. 8C) and the maximum frequency (Fig. 8B) also had a rather small dispersion. Again the distribution of the start frequency (Fig. 8A) was platycurtic, skewed towards higher frequencies and was most dispersed. The averaged power spectrum of 50 search pulses had a dominant frequency of 33.2 kHz and a -6 dB spectrum bandwidth of 1.6 kHz (Fig. 9). The bandwidth of the search pulses (Fig. 7C) varied more than in Nyctalus leisleri but, with 1.6 kHz, had the same peak value. The modal class and both classes on each side of it (range 0.8 to 2.4 kHz) contained only about 59% of the sample. Nine search pulses (T=11-18 ms) of the whole sample were of constant frequency, ın the range of 32.0-33.6 kHz. Table 4 shows the correlation coefficients of selected varıables of a random subsample of 49 search pulses of P. savın. The three strongest correlations indicate that shorter search pulses tended to have a higher start frequency and a larger sweep bandwidth. 288 PAESDAmSg 40 % 40 % A B 35| 35 30| 30 25| 25 N<100%) = 236 20| 20 N<100%) = 236 Median = 34,2 kHz 15 Median = 33.2 kHz 10 I Ü 28 32 36 40 44 start frequency (fs) kHz max. frequency (fmax) kHz 40 %& 40 4% C D 39 33 — 30 30 Fa) 23 H{i100%) = 236 H<100%) = 236 20, 20 Median = 32.8 kHz Hedian = 32.0 kHz 15| 15 10 iD 5| 5 0 28 32 36 +0 44 78 32 36 40 44 center frequency (fc) kHz end frequency (fe) kHz Fig. 8. Frequency varıables of search pulses of Pıpistrellus savı. Width of classes ıs 0.8 kHz. A: Start frequency at the beginning of a pulse. B: Instantaneous frequency at maximum signal amplitude. C: Center frequency at half the pulse duration. D: End frequency Fig. 9. Pipistrellus savı. Averaged power spectrum of 50 search pulses (settings cf. Fig. 3). Frequency span: 16-56 kHz. The vertical cursor is at the dominant frequency: 33.2 kHz. The -6 dB spectrum bandwidth is 1.6 kHz Search calls of echolocating Nyctalus leisleri and Pıipistrellus savıı 289 Fig. 10. Characteristic search pulse of Pipistrellus savüu. Oscillogram with superimposed frequency plot. BEN irhregtann, T=10.9 ms, fs=33.6 kHz, fmax=32.8 a eben kHz, fc=32.8 kHz, fe=32.0 kHz. The vertical cursor indicates the pulse 2 N center. The horizontal cursor is at 30 een te: wuse RUN s ei inet ggn ‘ Su steort N is laziyosapsaszratla eensslereNe kHz in the frequency plot Q3-Q1 95 % confidence interval of x Pulse duration T (ms) 3 2.9 12.1-12.7 Start frequency fs (kHz) 2.9 34.4-35.0 Maximum frequency fmax (kHz) 12 33.1-33.4 Center frequency fc (kHz) 12 32.6-32.9 End frequency fe (kHz) Ill 31.8-32.0 Bandwidth BW (kHz) 1.4 22 3.9 2.5 £ Qi and Q3 are the quartiles Q.o.>5, and Q.o.75, of the distributions. Q3-Q1 is the interval that contains 50 % of the whole sample (or 50 % ot the frequency distribution area) Table 4. Correlation coefficients (arranged from highest to lowest) of a selected variables of a random subsample of 49 search pulses of Pipistrellus savii eis, 10) 060 r(fmax, T)= -0.411** r(BW, fe) = +0.047 r(BW, T) = -0.514*** r(fc, T) = -0.357* r(fe, fs) = +0.451** Ede, 1)E 300: A goodness-of-fit test was consistent with normal distribution for all the variables. T: x’=4.8 (df Oel 3108 (di 6). imazı 45.3, (di=6). fc: ge -13%(df=4). fery2-23 (df>3), BW: 11.2 (df=6). Significance levels to reject the null hypothesis, Ho: p=0, are: ***P<0.001, **P<0.01 and &P<0.05. But P>0.1 für r(BW, fe) Discussion During the field work I realized that bats did show a different acoustic behaviour, depending on how they were released. If they were allowed to decide themselves when to fly away from the hand of the observer, most bats emitted short FM pulses with a steep frequency modulation. 290 P. E. Zıngg But when the bats were thrown upwards in the air, there was a good chance that they changed from the ınıtıal steep FM pulses to shallow modulated search pulses still within the range of the recording microphone. Generally pipistrelle bats (P. pipistrellus, P. nathusu, P. kuhli and P. savii) used to change quickly from approach pulses to search pulses and to turn round close to the place where they had been released. Sometimes they even started to hunt insects in the proximity ot the observer. Noctule bats (N. noctula and N. leisleri) on the other hand often used to fly straight away from the observer in fast flight and it was difficult to record search calls in this way. For this reason tape recordings of released and in their behaviour probably disturbed bats were not taken into account for the analysıs so far. To allow a comparison, parameter values of a pooled sample of 44 pulses recorded from 12 N. leisleri when they were thrown in the aır, ıs shown in Table 5. Table 5. Parameter values of echolocation pulses (N=44) of 12 Nyctalus leisleri, recorded shortly after the bats had been released Qi Median Q3 Q3-Q1 Minimum Maximum value value Pulse duration T (ms) Start frequency fs (kHz) 71 Maximum frequency fmax (kHz) 23) 26.8 72.) Center frequency fc (kHz) | | 26.2 2,5 End frequency fe (kHz) 23.9 25.1 2.1 Bandwidth BW (kHz) 33 73 5.8 6 Q1 and Q3 are the quartiles Q.o.25) and Qoz3 of the distributions. Q3-Q1 contains 50 % of the whole sample (or 50 % of the frequency distribution area) The parameter values of the pulse sample of released bats were generally similar to the parameter values of the search pulses of undisturbed, free tlying N. leislerı (compare with Tab. 1). The end frequency and the center frequency showed the least divergence. The extreme upper values of the start frequency and the bandwidth in Table 5 indicate that the sample contained a few pulses similar to those emitted in the approach phase. In general terms the varıables of search pulses analysed in this study varied less in P. savu than in N. leisleri. This points to a more regular flight behaviour in P. savız. I could not see Savı’s pipistrelles when they were hunting along the streets, since they used to fly higher than the bright streetlights at approximately 10 m above ground. The tlıght height of the Lesser noctules varıed considerably. On average the bats were hunting at the level of the street lights, but I observed Lesser noctules pursuing moths down to the ground or searching much higher than the street lights. This flexible tlıght behaviour may be the reason for the large dispersion of pulse interval durations. Search calls of P. savii had the lowest end- and center frequency of the four Pıpistrellus species (P. pipistrellus. P. nathusü, P. kuhli, P. savii) that occur in Switzerland (unpubl. data). The search calls of Savi’s pipistrelle appear to be closer to those of Eptesicus nılssoni than to those of the other three Pipistrellus species. This similarity between E. nılssoni and P. savii can be affırmed by their resembling appearance (e.g. colour of the skin, shape of Search calls of echolocating Nyctalus leisleri and Pipistrellus savü 231 the ear, projecting tail-tip), a fact already mentioned by MiıLLeEr (1912, p. 224). HORACEK and Hana (1986) now propose to place P. savır ın a separate genus Aypsugo. The correlations showed that in N. leisleri shorter pulses were generally of higher frequency, but without an increase of the bandwidth of the fundamental harmonic. It is conceivable that ın fact a higher sweep bandwidth was realized by introducing additional harmonics, which of course were not detectable by the system used ın this study. In general terms the values of the correlation coefficients are smaller in P. savii than in N. leisleri. The relations between bandwidth BW and pulse duration T were markedly different in the two species. In N. leisleri there was no significant correlation between these two variables in the case of search pulses. In P. savız on the other hand a significant negative correlation existed between the bandwidth and the pulse duration: When the search pulse duration decreased, their bandwidth increased. Since the end frequency showed only little varıability, the higher bandwidth of short pulses was due to an increased start frequency. This is affırmed by its negative correlation to the pulse duration. HooPrer (1969) and Watson (1970) evaluated the frequency range of N. leisleri by means of a heterodyning detector (Holgate Ultrasonic Receiver) only. HooPER (1969) determined 80 and 25 kHz as “upper and lower limits of the detected range of frequency sweep” and measured a pulse repetition rate of 8-10 pulses per sec (1.e. 125-100 ms ınterval duration) under indoor conditions (crawlıng). Only the lower frequency limit resembles the end frequencies of search calls presented in this paper. Watson (1970) gave an “approximate frequency range for field work” ot 15-70 kHz, which should be valid for both noctule species (N. noctula and N. leisleri). This frequency range is far wider than the range of search calls of N. leisleri (20-34 kHz) reported here. KOoNnsSTANTINOV and MAKAROV (1973) analysed 44 sıgnals of one P. savıi. The short pulse durations (0.8-2.3 ms) as well as the high start frequencies (90-100 kHz) and end frequencies (40-45 kHz) are consistent with indoor conditions. The authors mentioned a second harmonic and a near linear frequency modulation in the first part of the signal. PATLJAKEVITSCH (1980) measured a mean start frequency of 71.3 kHz and a mean end frequency of 35.7 kHz trom handheld P. savır. This ıs still above the highest end frequency I have found in search pulses. The pulse duration of his handheld bats ranged between 0.4 and 36 ms. When the bats were flyıng away from the wall of an avıary he measured pulse durations up to 5 ms. In our recordings the pulse durations were in the range of 8to 21 ms. The evidence of constant frequency (CF) pulses in N. leisleri and P. savı ıs not surprising. Pure CF-pulses have formerly been reported of N. noctula (PyE 1978, 1980; VOGLER and NEUWEILER 1983) and P. pipistrellus (PvE 1978, 1980; AHLEN 1981; MILLER and DEGn 1981). A sequential alternation of the end frequency on a pulse to pulse basıs, as known from N. noctula (AHLEn 1981; MıLLER and DEGN 1981; HARTLEY 1985), was rarely observed in N. leisleri. On the contrary, pulse sequences with end frequencies varyıng less than 1 kHz seem to be characteristic for this species. It is interesting to compare the positions of the modes of intervals between search pulses. In P. savız I got the equation ‘mode 2 = 1.53 x mode 1’ and in N. leisleri ‘mode 2 = 1.44 x mode 1’. A comparable relation between modes can be established from the interval distribution of search pulses of Nyctalus noctula (high flıght) in Anten (1981). For N. noctula HARTLEY (1985) measured a wing beat rate of sec (= 111 ms interval duration) in pursuit flight, corresponding to one pulse per wing beat. In cruising flıght he found two wing beats per pulse. The interpulse intervals (cruising flight) presented in this paper might be consistent with the hypothesis of two wıng beats per pulse (first mode) and three wing beats per pulse (second mode). Due to lacking behavioural data, the social calls described in this paper should be interpreted with caution. The fırst type of social call, recorded ın July 1986 at Meiringen, appeared to be an interaction between two individuals of N. leisleri. The first and the DL IE 7429 second type of social calls were also recorded when a P. pipistrellus passed the Lesser noctules. At the same time sequences of short pulses (3-4 ms, probably multiharmonie) with end frequencies of about 10 kHz (interpulse duration ca. 6 ms) were recorded. Figure 11 shows part of such a sequence enclosing an echolocation pulse. Fig. 11. Nyctalus leisleri. Oscillo- gram and superimposed period plot showing a search pulse (T=14.9 ms, fs=28.0 kHz, fc=24.8 kHz, fmax=fe=24.0 kHz) and part of a sequence of ten short pulses of social context (T= 2.7-4.2 ms, fe= ca. 10 kHz, inter- pulse duration = 4.6-8 ms). The vertical cursor indicates the pulse center of the search pulse. The total time from the start of the first short pulse to the end of the tenth short pulse was 113 ms In southern Switzerland the social calls resembling type 1 and 2 of Meiringen were recorded during the mating season (August 86) and there was no indication that interac- tions wıth other species occurred. On the basıs of the similar structure and the combined occurrence of call type 1 and 2 at Meiringen as well as in Southern Switzerland, I assume that these two types of social calls were actually emitted by individuals of Nyctalus leisleri at both recording sites. In the field, the social call of type 2 could be confused with echolocation pulses of Tadarida teniotis (ZBINDEN and ZınGG 1986). But signal analysis shows that search calls of T. tenıotis have not such a high start frequency and are not normally of linear period modulation. Acknowledgements I wish to thank ELISABETH BUCHER and RAPHAEL ARLETTAZ for their precious help in the fieldwork. Prof. V. Aerren kindly supplied me with Russian literature, parts of which were translated by Mrs. V. ScHnELL and R. Hope. I thank Dr. ©. BERNATH, Departement of Audiology of the Berne University Hospital, for permitting the use of the FFT-analyser. I am greatly indebted to Dr. KArı ZBINDEN for critically reading the manuscript, making important suggestions for its improvement and correcting the text. This study was supported by the Swiss National Science Foundation (grant no. 3.177-0.85), Zusammenfassung Suchflugortungslaute von Nyctalns leisleri und Pipistrellus savıı (Mammaha: Chiroptera) in der Schweiz Suchflugortungslaute jagender Nyctalus leisleri und Pipistrellus saviı wurden mittels QMC S100 Detektor und 8:1 Frequenzteiler auf der schweizerischen Alpennord- und Alpensüdseite registriert. Bei beiden Arten wurden sowohl schwach frequenzmodulierte als auch konstantfrequente Suchtlug- laute festgestellt. Für 204 analysierte Suchfluglaute von N. leisleri waren folgende Medianwerte charakteristisch: Sıgnaldauer 9.2 ms, Anfangsfrequenz 25.1 kHz, Zentrumsfrequenz 24.3 kHz, Endfrequenz 23.7 kHz, Sıgnal-Bandbreite 1.6 kHz. Die Intervalldauer (N=2153) zwischen den Signalen zeigte eine bimodale Verteilung, mit einem starken Modus bei 243 ms und einem zweiten, schwächeren bei 350 ms. Für 236 Suchfluglaute von P. savıi waren folgende Medianwerte typisch: Signaldauer 12.1 ms, Anfangsfrequenz 34.2 kHz, Zentrumsfrequenz 32.8 kHz, Endfrequenz 32.0 kHz, Signal-Bandbreite 2.2 kHz. Die Intervalldauer (N=1341) zwischen den Signalen war bimodal verteilt, mıt einem starken Modus bei 192 ms und einem zweiten, schwächeren bei 294 ms. Bei beiden Spezies wurden zwischen verschiedenen Variablen der Suchfluglaute signifikante Korrelationen ermittelt. Search calls of echolocating Nyctalus leisleri and Pipistrellus savü 293 Zusätzlich werden als Soziallaute interpretierte Ruftypen von N. leisleri beschrieben. Der erste Typ bestand aus einer Gruppe von vier Lauten im Frequenzbereich von 38-16 kHz. Der zweite Typ, ein linear periodenmodulierter Einzellaut (T=20-27 ms), überstreicht ein Frequenzband von 21-10 kHz und kann ım Felde mit Ortungsrufen von Tadarıda teniotis verwechselt werden. References AHLEN, 1. (1981): Identification of Scandinavian Bats by their sounds. Swed. Univ. of Agricult. Scı. Dept. Wildlife Ecol., Report 6, Uppsala. FENTonN, M. B.; BErL, G. P. (1981): Recognition of Species of Insectivorous Bats by their Echoloca- tion Calls. J. Mammalogy 62, 233-243. HARTLEY, D. J. (1985): Analysis of the echolocation behaviour of some British bats recorded in the field. Ph. D. Thesis, Polytechnic of Central London. Queen Mary Coll., Univ. London. Hooper, J. H. D. (1969): Potential use of a portable ultrasonic receiver for the field identification of flying bats. Ultrasonics 7, 177-181. HOoRACER, 1.; HANAK, V. (1986): Generic status of Pıipistrellus savii and comments on classification of the genus Pipistrellus (Chiroptera, Vespertilionidae). Myotis 23-24, 9-16. KoNSTANTINOV, A. I.; MAKARoOv, A. K. (1973): Echolocating signals of some native species of bats (Vespertilionidae). In: Problems of comparative physiology of analysers. Vol 3: Echolocation ın bats. Ed. by E. Sh. Airapetianz. Leningrad: Publishing House of the State Leningrad University. 2944. (In Russian). MILLER, G. S. (1912): Catalogue of the Mammals of Western Europe (Europe exclusive of Russia) in the collection of the British Museum. London: British Museum (N.H.). MILLER, L. A.; Decn, H. J. (1981): The Acoustic Behavior of Four Species of Vespertilionid Bats Studied in the Field. J. Comp. Physiol. A 142, 67-74. PATLJAKEVITSCH, N. D. (1980): Echolocation signals of Vespertilionidae. In: Rukokrylyje (Chiropt- era). Ed. by V. E. Sokolov. Moscow: Nauka. 213-270. (In Russian). Pv£, J. D. (1978): Some preliminary observations on flexible echolocation systems. In: Proc. Fourth Internat. Bat Research Conf. Ed. by R. J. Olembo, J. B. Castelino and F. A. Mutere. Nairobi: Kenya Literature Bureau. 127-136. PyeE, J. D. (1980): Adaptiveness of echolocation signals in bats. Flexibility in behaviour and in evolution. Trends in NeuroSciences (TINS), October 1980, 232-235. Sachs, L. (1978): Angewandte Statistik. Berlin, Heidelberg, New York: Springer-Verlag. TUPINIER, Y.; BIRAUD, Y.; CHIoLLAZz, M.; EscuDIE, B. (1980): Analysıs of vespertilionid sonar signals during cruise, pursuit, and prey capture. In: Proceedings of the Fifth International Bat Research Conference. Ed. by D. E. Wilson and A. L. Gardner. Lubbock (Texas): Texas Tech. Univ. Press. 29-38. VOGLER, B.; NEUWEILER, G. (1983): Echolocation in the noctule (Nyctalus noctula) and horseshoe bat (Rhinolophus ferrumequinum). J. Comp. Physiol. A 152, 421-432. Watson, A. (1970): Electronic aids to the ıdentification of bats in flıght and to their study under natural conditions. Bijdragen tot de Dierkunde 40, 99-102. ZBINDEN, K.; ZınGs, P. E. (1986): Search and hunting signals of echolocating European free-tailed bats, Tadarida teniotis, in southern Switzerland. Mammalıa 50, 9-25. Author’s address: PETER E. ZınGs, Institute of Zoology, University of Berne, Baltzerstrasse 3, CH- 3012 Berne, Switzerland Z. Säugetierkunde 53 (1988) 294-300 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 The development of visual acuity in treeshrews (Tupaia belangeri) By BırGIT HERTENSTEIN, ELKE ZIMMERMANN and H. RAHMANN Institute of Zoology, University Hohenheim, Stuttgart Receipt of Ms. 1. 6. 1987 Abstract Ontogenetic development of visual acuity was tested in 14 young treeshrews (Tupaia belanger:) from the moment of eye opening (day 16 to 23) until leaving their nest (day 30 to 34), using an optomotoric drum. First optokinetic nystagmus (OKN) was found between day 4 and 5 after eye opening with individual values of visual acuity varying considerably between 34,5 and 370 minutes of arc. Before leaving the nest all values were about 11 minutes. In general, the longer the process of eye opening took (from a few hours to three days) the later was the first OKN. No correlation was found between the onset of eye opening and the first OKN. There were no differences in the development of visual acuity between the two sexes. In addition to the optomotoric drum training, visual acuity of two adult male tupaias was determined using a visual discrimination choice apparatus. According to all tests, visual acuity of adult treeshrews was found to be 10 and 8,6 minutes respectively. The extent of visual acuity found in this study was compared with that of some primates. Introduction Currently the Tupaiidae are considered to be the group of recent mammals that is most sımilar to the ancestors of primates (STARCK 1978). However, despite numerous mor- phological, physiological and ethological investigations, there has been no final decision as to their relationship to other mammalıan taxa (LUCKETT 1980). Therefore, they have been placed in a separate mammalıan order, Scandentia. The diurnal and semiarboreal tupaias live in southeast Asıa, where males occupy overlapping home ranges with females (KawamiıcHi and KawamichHiı 1979). They feed on different sorts of fruit, insects and even some smaller vertebrates. Their senses of smell and hearing as well as their vision are highly developed. Visual acuity of adult tupaias has already been determined with visual discrimination training but with different results. Thus, OrpDy and SAMORAJSsKI (1968) have found a visual angle of 0,5 to 1,5 min of arc and SCHÄFER (1969) has reported one of 6,8 mın of arc. Even though research on the visual acuity of the eye in various groups of anımals has been reported in the past and with many different methods (RaAHMmAnN 1967; BOOTHE et al. 1985; for review), so far there is no information on the ontogenetic development of vision in Tupaia belangeri. The purpose of this study is to show how visual acuity develops in young tupaias from the moment of eye opening until after they leave their nests, and thus, to provide a basıs for comparison with the visual development of various primates. Material and methods Determination of visual acuity during development of tupaias by means of the optomotoric nystagmus (OKN) 14 young tupaias born at the breeding colony at the Department of Zoology, University of Stuttgart Hohenheim, were tested. Housing and maintenance of breeding pairs and postnatal behavioral development of young has been described previously (HERTENSTEIN et al. 1987). Tupaias are born U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5305-0294 $ 02.50/0 The development of visual acnity in treeshrews 235 blind. Data collection began at eye opening about 19 days after birth, and continued another 10 to 12 days until they had left their nest box for the first time. At this time tupaias develop a highly coordinated locomotor ability to move through their semi-arboreal environment and react promptly to potential dangers. 5lem 2 Fıg. 1. Optomotoric drum for determination of visual acuity of young treeshrews. (w = white sheet, used to reduce the visual field \ of the anımal) 6lcm The experiments were conducted in an optomotoric drum having a height of 51 cm and a diameter of 64 cm, with vertical black and white stripes of equal width (Fig. 1). There were 15 different stripe- patterns each of them having a particular width of stripes ranging from 12 mm to 0,4 mm. A 75W bulb, hanging over the centre of the barrel, provided equal illumination. There were no windows or other light in the room. The animals were found to remain calm if held by hand while being tested. Visual acuity was tested once each day in the morning. The young were taken out of the parental cage within their nestboxes. The anımals were allowed to adapt to the light for 15 to 20 min. E Testing started with the pattern with the - widest stripes (d = 12 mm) in a short distance ( (E = 10 cm, see Fig. 2). The distance E was d‘ I ae increased, ıf the anımal showed a clear op- \ tomotoric nystagmus (OKN). The next Std Eye smaller stripe pattern was used once the dis- p tance between head and pattern could not be ER increased any further (diameter of the barrel Sz2dhlemE2 degrees = 32 cm). Visual acuity was calculated from VISUAL ANGLE «= ft» Efcm) |[ofarc the exact field of sight (E) at which, for a definite width of stripes (d), the animals still Fig. 2. Formula for calculating visual acuity. (See text responded with an OKN, using the formula for details) inEie.22: As the attention of the anımals decreased rapidly, they tried to leave the hand or simply went to sleep. Therefore many rests were necessary during each experiment and in some cases the test had to be broken off completely. Determination of visual acuity in adult tupaias by means of visual discrimination tests The tests in this case were run with two adult male tupaias of an age of 42 and 7% months respectively, both of which had been reared by hand (HERTENSTEIN et al. 1987). The animals were kept individually in wooden cages of a size of 0,5 m’, and containing branches for climbing, shelves for sitting upon and a sleeping box. The ground was covered by fine sawdust and straw. Their food consisted of tupaia-pellets (by Altromin) and water ad lıb., in addition to fruit, sunflower seeds, nuts, oatmeal, egg and cat food given ın rotation. Insects were not used ın the diet, since mealworms (Tenebrio larva) were used as rewards in the discrimination tests. The “two alternative discrimination apparatus” utilized was similar to training systems used to test visual acuity in golden hamsters (Mesocricetus auratus), lemmings (Lemmus lemmus, Myopus schis- tıcolor) and various species of american deermice (Peromyscus) (RAHMANN 1961; RAHMAnN and EssER 1965; RAHMANnN and RAHMAnN 1966; RAHMANN et al. 1968). From a start box (Fig. 3) the anımals enter the choice box where they have to decide between one of two patterned doors at the end of the choice box, separated by a partıtion. The patterns for the discrimination experiments were fixed to the doors. The patterns could be changed quickly from one side to the other by pulling strings. If the anımal chose the correct door, he received a peace of mealworm as a reward. The anımals made their 365 VISUAL ANGLE (MIN. OF ARC) 296 40 cm AWARD-BOXES De 1 Al f T IL | PATTERN |DOORS J / t m | =! | | = FE |S =) eier | = | | | a H 1 - | | CHOICE BOX | | &| E S| START BOX S a | a | | | | I 1 t FE (02) Fig. 3. Discerimination apparatus for determination of visual acuity of adult treeshrews. Pattern doors were manually alternated from left to right by pulling strings (s) else) --------- „-oN FIRST OKN (DAYS AFTER Birgit Hertenstein, Elke Zimmermann and H. Rahmann way back to the start box through return alleys, and the next test was started immediately. The whole apparatus was covered by a lid of plexiglas with slots for ventilation and presentation of rewards. A 75W bulb was hung just above the apparatus to illuminate the area around the patterned doors. The tests were run once a day in the morning. The anımals had to decide between a pattern of black and white vertical stripes of equal width (positive pattern, with reward) and a grey surface of the same brıghtness (negative pattern, door locked). The pattern was offered both on the right and on the left door, 20 times each, in a predeterminal, irregular order. There were 14 different patterns with varıous wıdths of stripes from 12 mm to 0,6 mm. The distance between the choice box and the pattern doors was 48 cm. The criterion of significance for the number of correct choices in 40 runs was fixed at 70%, for a 99% probability (KorrEr 1969). If the results of a test were not significant after an obvious lack of concentration of the test anımal, the test was repeated with the same width of stripes on the following day. Quite often the result was then significant. To avoid the possibility that the anımals were choosing different degrees of brightness of the patterns rather than deciding between stripe pattern and plain grey, a second series of tests was run, in which a choice had to be made between a vertical stripe pattern (positive pattern) and a horizontal one. All the other conditions of the test remained the same. Subsequently visual acuity was determined for a shorter choice distance of 27 cm, in order to have a comparison between greater and shorter choice distances. 1 5 9 DAYS AFTER EYE OPENING Fig. 4. Development of visual acuity in 9 male and 5 female treeshrews. (For youngsters from the same litter identical symbols are used) p 36o. Uta IE n< | X | N) R | 1 zZ = MM) ©) | | u — > 1 ‚a > => > = —zleo> zZ © — u |zZ2 } ©) le M 84182 O > OL 2 h Koraif Seisy wo = <_ u WuOOo | BI ı < 1 _ un > 13 1 5 9 13 DAYS AFTER EYE OPENING The development of visnal acnıty in treeshrews 297 Results Development of visual acuity during development of Tupaia belangeri As shown in fig. 4a and b, four anımals responded with a clear optomotoric reaction on the first day after eye opening. A first reaction was shown by three animals on the second day after eye opening, three on the third day, two on the fourth and one on the fifth. Within this period the individual visual angle varıed considerably (34,5 min to 370 min = 6,2°). During the following eight days the range of the values decreased and, by the time the animals left their nests, they had reached a similar level of 11#0.75 minutes in the males and of 11,7+2,68 minutes in the females. The great interindividual differences in the first occurence of an OKN and the values for visual acuıty might be caused not only by different visual efficiency, but also by the stress on particularly sensitive anımals when being handled. When their eyes started to open the test anımals were between 18 and 22 days old. Separation of the eyelids took as long as three days but, in other cases, just a few hours. Notwithstanding how long the process of the eye opening took, the start of the process was counted as day number one. The first optomotoric response tended to be delayed when the process of eye opening took longer (see tables ın Fig. 4a and b). No correlation between the age of the test anımals at the start of the eye opening and the first OKN was found. Visual acuity of adult tupaias First, the anımals were allowed to become familiar wıth the discrimination choice apparatus. They were rewarded if they used the striped door by chance (width of stripes in this case 12 mm). On the second test day test anımal H (V’Ty;) already chose the striped door sıgnificantly more often than the plain grey door. Test anımal W (VTw), which was only 4/2 months old, seemed nervous during the daily 5 to 10 minutes in the apparatus and took 7 days before he chose significantly between the two alternatives. To determine visual acuity (minimum separabile = perception of separation between stripes), the width of stripes was gradually reduced, from 7 to 0,6 mm, during the following days of testing, whereas the length of the partition was 48 cm. The door with the negative ve 100 pattern remained locked. An attempt z to go down the wrong alley for more = than one head-length was scored at a ae miss. VTjı chose significantly between = stripes and plain grey down to a width of stripes of 1,4 mm (see Fig. 5). This is equivalent to a visual angle of 10 minutes of arc. VI’y discriminated a 2 stripe width of 12 mm, equivalenttoa pas el on Baumes Key) Eon visual angle of 8,6 minutes of arc. strip wIDTH [m] : 70 70 50 30 18 Ik u 12 ın 08 06 Visual acuity for choosing between yjsyaL ancıe horizontal and vertical stripes was fmnures or arc] 11,5 AELEIIUES for VTy and 10 for V’Tiw. Fıg. 5. Number of correct trials after the gradual This is virtually the same as was found „eduction of the stripe width. (Day one means the first in the first set of experiments. As the training day after learning the pattern discrimination) partition and thus the distance E was reduced from 48 to 27 cm VTw did not show any change ot efficiency ın hıs visual acuıty. For the stripe width of 0,8 mm, which VTw could still discriminate, the visual angle was 10,2 minutes. VI}, on the other hand, tended to choose one particular side increasıngly as the experiment proceeded. This made a determination of the smallest visual angle impossible. 50 :501 501 358 215 129 115 100 8 72 57 43 298 Birgit Hertenstein, Elke Zimmermann and H. Rahmann There were great differences in how the anımals behaved in the apparatus. Sometimes they were calm and attentive to the apparatus. In other instances, the test anımal dashed around the apparatus ın a hectic manner. An anımal might rush into one of the choice alleys, realize his error, and then quickly turn around and go into the correct alley. A run of this sort was scored as a miss, even though after an initial error at the partition, the anımal definitely made the right choice. Discussion Many physiological as well as psychological factors influence determination of visual acuity. Thus results of this study might not be directly comparable with the results of previous authors or investigations into other species. Methods of investigating visual acuity often vary considerably from author to author. Different intensities of illumination can produce differences in the results, as could be shown by CavoniIus and Rossıns (1973) with rhesus J00 I =TREESSHIREWY monkeys and RaHmann et al. € Bu -BUSHBABY (1968) with deermice (Peromys-- “ cus). Moreover, the willingness S P =PIG-TAILED MACAQUE and ability to concentrate shown 80 R =RHESUS MONKEY by the test anımals plays a roe Ba -BABOON (RAHMAnN 1967). Thus, the failure 2 6 =GIBBON to perform, which occurred again = O =ORANG -UTAN H =HUMAN [0}) >) and again in this study, both in the trainıng apparatus and in the drum, might be the result of many factors including distraction, weariness, and stress of the test anımals. The value for visual acuity of 8,6 minutes of arc, which has been obtained with the adult tupaias in 20 our training apparatus, ıs similar to the results of SCHÄFER (1969) found in a comparable discrimina- tion apparatus for tupaias (visual acuıty: 6,8 minutes of arc). Using a “noncorrection method of train- 40 De 1 3 5 Y 9 1 WEEKS AFTER EYE-OPENING ing”, Orpy and SAMORAJsKI Fig. 6. Development of visual acuity in tupaias in compari- (1968) found a visual acuity Of son to that of some primates. (Treeshrews and bushbaby tupaias of 0,5 to 1,5 minutes of arc from own data, other data from Fogzs and Kına 1982) which appears quite high (Tab. 1). | Figure 6 shows the time course of the development of visual acuity in tupaias compared with that of some primates: tupaias and rhesus monkeys achieve some 14 minutes at an age of about one week after eye opening (rhesus and other primates: eyes are open at bırth). Baboons show this value soon after birth (Foses and Kınc 1982). Pig-tailed macaques (Macaca nemestrina) pass the limit of 14 minutes after 2 to 3 weeks, gibbons after 3 weeks, orangs after 5 weeks and man after about 11 weeks (Foges and KınG 1982). A bushbaby (Galago senegalensis; behaviourly investigated by ZIMMERMANN 1987) tested with the same method, showed a visual acuity of 18,7 minutes after 25 days (end of the experiment). Obviously with increasing phylogenetic level, the speed of development of visual acuity slows down. This is consistant with the observation that the “higher” the phylogenetic The development of visual acuity in treeshrews 299 Table 1. Visual acuity of treeshrews tested by different authors This work Schäfer (1969) Ordy and Samorajski (1968) Visual acuıty 6-8 Oesztonle 5 (minutes of arc) Apparatus Two-alternative choice Two-alternativechoice Two-alternative choice apparatus apparatus apparatus Method With reward, no punish- With reward, with pun- With reward, a „non-cor- ment ishment (sudden noise) rection method of train- cc ing rank of a primate is, the longer its ontogeny takes. Tupaias, that independently leave Table 2. Visual acuity of treeshrews and some their nest some 12 days after their eyes primates opened need a good visual acuity at that Treeshrews this work, other data from FoBes time to find their way through the environ- and Kıng (1982) ment, to catch small insects and to recog- nıze dangers. The later an anımal reaches Species Visual acuity (minutes) independence, the less urgent is the need for rapıd development of visual acuity. A Treeshrews human child, for example, only reaches the Common marmosets Fr Ceb k efficiency of an adult at an age of five years u elnenkers (BOOTHE et al. 1985). Olive baboons In comparing the values for visual acuity Pig-tailed macaques between treeshrews and primates (Tab. 2), Er menker h h b a Chimpanzees the primates show a better visual acuity Human beings than the measured acuity of tupaias (about 9 minutes of arc). IN oooWNDOoOoo—m Acknowledgements Thanks are due to Dr. $. K. BEARDER, Anthropology Section, Oxford, and to Prof. Dr. B. BEnsoN for checking the English style and for critical reading of an earlier version of this manuscript, to Prof. Dr. H. SprAankEL, Department of Neuropathology, Gießen, for providing two pairs of Tupaia belangeri for founding the breeding stock of our University, to Dipl.-Ing. (FH) H. Oster, IBM Deutschland, for making the stripe patterns, to W. KUTTRITZ, TH. PLÖTTNER and G. TOLKMITT for assisting in constructing the apparatus for the experiments and to W. TRUMPP, G. ZANDER and T. KLescH for their permanent and indulgent support in maintaining the treeshrews in the anımal house at Stuttgart-Hohenheim University. Zusammenfassung Die Entwicklung der Sehschärfe bei Spitzhörnchen (Tupaia belangeri) Die ontogenetische Entwicklung der Sehschärfe wurde an 14 jungen Spitzhörnchen (Tupaia belanger:) ab dem Zeitpunkt des Augenöffnens (16. bis 23. Tag) bis zum Verlassen des Nests (30. bis 34. Tag) mit Hilfe einer optomotorischen Trommel getestet. Der erste optokinetische Nystagmus (OKN) trat zwischen dem 1. und 5. Tag nach dem Augenöftnen auf, dabeı streuten die individuellen Sehschärfe- werte beträchtlich (34,5 bis 370 Bogenminuten). Bis zum Verlassen des Nests pendelten sie sich auf etwa 11 Bogenminuten ein. Es zeigte sich eine Tendenz, daß mit zunehmender Dauer des Augenöft- nens (wenige Stunden bis 3 Tage) eine Verzögerung des ersten Auftretens eines OKN einhergeht. Zwischen den Geschlechtern konnte in bezug auf die Sehschärfe kein Unterschied festgestellt werden. Zusätzlich wurde die Sehschärfe zweier adulter Tupaia-Männchen in einer 2-fach-Wahlapparatur bestimmt. Die ermittelte Sehschärfe lag bei 10 bzw. 8,6 Bogenminuten. Die Sehschärfewerte wurden mit denen einiger Primaten verglichen. 300 Birgit Hertenstein, Elke Zimmermann and H. Rahmann References BOOTHE, R. G.; Dogson, V.; TELLER, D. Y. (1985): Postnatal development of vision in human and nonhuman primates. Ann. Rev. Neuroscı. 8, 495-545. Cavonıus, C. R.; RoBBIns, D. O. (1973): Relationships between luminance and visual acuity in the Rhesus monkey. J. Physiol. 232, 239-246. Foßges, J. L.; Kıng, J. E. (1982): Primate Behavior. New York, London: Academic Press. HERTENSTEIN, B.; ZIMMERMANN, E.; RAHMANNn, H. (1987): Zur Reproduktion und ontogenetischen Entwicklung beı Spitzhörnchen (Tupaia belangeri). Z. des Kölner Zoo 30 (4) 119-133. KAawamichHı, T.; KAwaMmIcHı, M. (1979): Spatial organization and territory of tree shrews (Tupaia glis). Anım. Behav. 28, 381-393. KOLLER, $. (1969): Neue graphische Tafeln zur Beurteilung statistischer Zahlen. Darmstadt: Stein- kopff Verlag. LuckEtT, W. P. (1980): Comparative Biology and Evolutionary Relationships of Tree shrews. New York: Plenum Press. Orpy, J. M.; SamoRAJsKı, T. (1968): Visual acuity and ERG-CFF in relation to the morphologic organization of the retina among diurnal and nocturnal primates. Vision Res. 8, 1205-1225. RAHMAnn, H. (1961): Einfluß des Pervitins auf Gedächtnisleistungen, Verhaltensweisen und einige physiologische Funktionen von Goldhamstern. Pflügers Arch. ges. Physiol. 273, 247-263. — (1967): Die Sehschärfe bei Wirbeltieren. Naturwiss. Rundschau 20, 8-14. — (1969). Sehschärfe. Bild der Wissenschaft 6, 648-655. RAHMAnn, H.; Esser, M. (1965): Bestimmung der Sehschärfe (Minimum separabile) sowie Dressur- verhalten des skandinavischen Berglemmings (Lemmus lemmus L.). Z. Säugetierkunde 30, 47-53. RAHMaAnn, H.; RAHMann, M. (1966): Sehschärfe (Minimum separabile), Dressurverhalten und vergleichende Augenlinsenmessungen beim Waldlemming (Myopus schisticolor Lillj.) und Berg- lemming (Lemmus lemmus L.). Z. Säugetierkunde 31, 396-402. RAHMAnN, H.; RAHMANN, M.; Kıng, J. A. (1968): Comparative visual acuity (Minimum separabile) in five species and subspecies of deermice (Peromyscus). Physiol. Zool. 41, 298-312. SCHÄFER, D. (1969): Untersuchungen zur Sehphysiologie des Spitzhörnchens Tupaia glis (Diard, 1820). Z. vergl. Physiol. 63, 204-226. STARCK, D. (1978): Vergleichende Anatomie der Wirbeltiere auf evolutionsbiologischer Grundlage. Berlin: Springer Verlag. 1, pp 189. ZIMMERMANN, FE. (1988): Aspects of reproduction, behavioural and vocal development in Senegal bushbabies (Galago senegalensis). Int. J. Primatol. (in press). Authors’ address: BIRGIT HERTENSTEIN, Dr. ELKE ZIMMERMANN, Prof. Dr. H. RAHMAnn, Institut für Zoologie, Universität Stuttgart-Hohenheim, Garbenstraße 30, D-7000 Stutt- gart 70 Z. Säugetierkunde 53 (1988) 301-316 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Mecanismes de coexistence dans une guilde de murides insulaires (Rattus rattus L., Apodemus sylvaticus L. et Mus musculus domesticus Rutty) en Corse: Consequences Evolutives Par L. GRANJON et G. CHEYLAN Laboratoire d’Eco-Ethologıe, Institut des Sciences de !’Evolution, Universite des Sciences et Techniques du Languedoc, Montpellier, France Reception du Ms. 02. 04. 1987 Abstract Mechanisms of coexistence in a guıld of insular Murıids (Rattus rattus L., Apodemus sylvaticus L., and Mus musculus domesticus Rutty) in Corsica: Evolutionary consequences Studied the guild of Murids from the mediterranean island of Corsica (Rattus rattns, Apodemus sylvaticus and Mus musculus domesticus), in order to elucidate the relationships between its members and to propose potential evolutionary trends for them. The distribution of the three species at a regional scale is apprehended through extensive trapping throughout Corsica. The density fluctua- tions and the spatıal distributions at a local scale are defined from a capture-recapture program in a North-Western locality of the island. Finally, the interindividual relationships between members of the three species are studied in the laboratory. Results indicate that the distribution and demography of the wood mouse is affected by the presence of the two other Murids, as are the activity patterns in captivity of both A. sylvaticus and M. musculus by the presence of R. rattus ın captivity. Nevertheless, the coexistence of the three species is made possible by a shift of the demographic cycle of the wood mouse as well as of its habitat preference. This set of data suggests that the wood mouse, which was the first among the three species here studied to colonize Corsica, is also the one which suffers the most from interspecific interactions. This situation can lead A. sylvaticus to differentiate more rapidly, according to the “taxon” cycle theory of Mac ARTHUR and Wırson (1967). Introduction A la suite des travaux classıques de Mac ARTHUR et Wırson (1963, 1967), l’etude de la biogeographie insulaire s’est focalısee sur l’Equilibre colonisations/extinctions. Pourtant, dans leur ouvrage paru en 1967, ces auteurs ont consacr@ un chapitre entier A l’evolution des populations insulaires («Evolutionary changes following colonization»), aspect par ailleurs largement discute dans les ouvrages de WILLıaMSoN (1981) et de BROWN et GIBSON (1983). Constatant que les colonisateurs se recrutent essentiellement parmıi les especes habıtant les milieux secondaires (milieux ouverts et instables, provenant generalement de la degrada- tion des milieux primaires, ı.e. forestiers), Mac ARTHUR et Wırson (1967), reprenant les travaux de Wırson (1961) et de CarLouıst (1966) sur les fourmis et les Composees du Pacifique, definissent un «taxon cycle» durant lequel l’evolution des communautes insu- laires passe par trois phases: 1. colonısation des milieux secondaires par des especes strateges «r>; 2. penetration des milieux forestiers par ces especes, entrainant une perte du pouvoir de dispersion et un glissement vers une strategie de type «K»; 3. differenciation aboutissant Eventuellement A une speciation souvent accompagnee d’une distribution relique. Selon les cas, une recolonisation des milieux ouverts peut suivre une radiation adaptative, A moins que de nouveaux colonisateurs ne s’y soient deja installes. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5305-0301 $ 02.50/0 302 L. Granjon et G. Cheylan Ce schema a £te par la suite appliqu& avec succes aux avıfaunes des Antilles par RICKLEFS (1970) et RıckLers et Cox (1972, 1978). Il insiste sur l’ımportance de la competition interspecifique, qui repousse dans les milieux forestiers les colonisateurs primitifs, renfor- cant l’isolement de ces populations et leur evolution distincte (MAC ARTHUR et WILSON 1967: 157). Brown et GiBson (1983) donnent plusieurs exemples empruntes aux oiseaux et aux lezards ıllustrant, selon eux, l’ımportance de la competition interspecifique dans la distribution d’esp£ces ınsulaires. Bien que l’importance m&me de la competition ait Eet@ recemment remise en cause (ConNELL 1983; SCHOENER 1983; BRADLEY et BRADLEY 1985), les nombreux travaux r&alıses sur des especes du genre Anolıs aux Antilles illustrent de fagon pertinente la realıte de ce phenomene dans les groupes composes d’especes affınes, chez les Sauriens par exemple (PacALA et ROUGHGARDEN 1985; RUMMEL et ROUGHGARDEN 1985). Les quelques analyses realisees sur les communautes de Rongeurs aboutissent aux me&mes conclusions (GRANT 1970; CROWELL 1973, 1983; CROWELL et PımM 1976; HALLET 1982; HALLET et al. 1983; DuEsER et PORTER 1986), tout en mettant en @vidence des differences selon les especes etudiees. D’une facon generale, la competition est d’autant plus intense que les especes sont Ecologiquement, ou phylogeniquement, proches, ce qui confirme les predictions du modele theorique (MAc ARTHUR 1972). La communaute& de Rongeurs Murides de Corse se prete bien & l’etude de ces processus evolutifs suivant la colonisation d’un milieu nouveau et des processus de competition pouvant y Etre associes. Connaissant approximativement les modalıtes d’arrıvee de ces especes en Corse (VIGNE 1983a et b), nous avons tente de reconstituer le schema de mise en place puis d’evolution de la guilde formee par le rat noir (Rattus rattus), la souris domestique (Mus musculus domesticus) et le mulot sylvestre, (Apodemus sylvaticus), A travers leur repartition actuelle dans /’ile, leurs caracteristigues demographiques et spatiales au niveau d’une station, et enfin leurs relations interspecifiques en conditions de labora- toire. Materiel et methodes Analyse Ecologique Dans le but de connaitre la r£partition des trois Murides en Corse, 22 stations de piegeage r&parties sur tous les Etages bioclimatiques de l’ile entre O et 2064 m ont Et& Echantillonnees de 1981 a 1985. La majorite de ces piegeages a te realisee a l’aide de pieges grillages de type Firobind. Les resultats permettent de pr£ciser la repartition altitudinale des trois especes de Murides. Par ailleurs, un quadrat de piegeage a Ete installe dans le vallon d’Elbo, sıtue dans la reserve naturelle de Scandola, au nord-ouest de la Corse (47,08 gr lat N; 6,93 gr longE). Ce quadrat comporte 94 jalons (76 en janvier 84), avec une maille de 20 me£tres, l’ensemble couvrant une superficie de 3,76 ha (3,04 ha en janvier 84). Six sessions de capture ont Et€ menees, en janvier, avril, juillet, octobre 1984, et en mars et juillet 1985. Durant ces periodes, les Rongeurs £taient pieges pendant 3 nuits successives, puis une ce nuit de capture, 3 jours plus tard, permettait d’estimer les densites de population gräce a l’indice de Lincoln. Deux pieges Etaient installes & chaque jalon durant l’annee 1984, un seul lors des deux sessions de 1985. Les pieges Etaient appätes A l’aıde d’une päte a base de farine, de sardines et d’huile d’olive, melange qui s’est revele attractif pour les trois especes considerees icı. Les anımaux captures etaient marque&s (amputation de phalanges ou bagues nume£rotees a l’oreille), sexes, peses et leur Etat reproducteur etait note. Ils Etaient ensuite reläches A leur jalon de capture. Les densites de population ont Et estimees par l’indice de Lincoln, afın de faciliter les comparai- sons avec la plupart des autres etudes (Orsını 1982; CassaınG et CRosET 1985). Parallölement, un releve de vegetation a ete effectug, en notant autour de chaque jalon la presence et Pabondance des especes vegetales presentes. Cing zones ont pu alors Etre definies (Fig. 1). Ce sont: Zone A: Friches & Graminees et petites Papilionacees, couvrant environ 1 ha disperse en trois champs entoures de murs de pierres seches. / nula viscosa, Ferula communis, Euphorbia helioscopa et divers Plantago en sont caracteristiques. Zone B: Zones degradees en adret sur sol rocheux, & Cistus monspeliensis et Polygonum scoparium, avec touffes de Pistacıa lentiscus. Mecanismes de coexistence dans une guilde de murides insulaires 303 Zone C: Maquis (Oleo-lentiscetum climacique) ä Erica arborea, C. monspelensis, P. lentiscus, Myrtus communis, Olea europaea, Phillyrea latifoha et Ph. angustifolia. Zone D: Maquis Eleve a Arbutus unedo et E. arborea avec Ph. latifoha et Viburnum tinus (hauteur du toıt de la formation: 2 m) Zone E: Ripisylve dense & P. lentiscus avec Ph. latifolia et V. tinus, atteignant 2&3 m de haut. On y observe deux especes d’arbres caracteristiques des ripisylves mediterraneennes chaudes: Fraxinus ornus et Vitex agnus castus; cette formation comporte par ailleurs de nombreuses lianes: Rubus fruticosus et ulmifolois, Smilax aspera, Tamus communis, Clematis flammula et vitalba. —.- Eee ARRERZ 20m . > Fıg. 1. Plan du quadrat de piegeage d’Elbo et des zones de vegetation recensees Les effectifs de capture de chaque espece dans ces cing milieux au cours des six sessions de piegeage ont permis de r&aliser une analyse factorielle des correspondances afın de preciser la r&partition spatio- temporelle de ces trois especes sur le quadrat. Ensuite, la repartition de chaque espece dans chacune de ces cing zones a Et€ exprimee en pourcentage du nombre total de captures de cette espece au cours d’une periode donnee. Les pourcentages de capture ainsi obtenus ont Et€ rapportes A la surface occupee par ces zones de vegetation exprimee en pourcentage de la surface totale du quadrat. Cet indice de repartition. N % capture sp N dans milieu M JE = % S_occup€ par M a et€ calcul€ pour chaque espece dans chaque milieu, d’une part lors de chacune des 6 sessions de piegeage, et d’autre part A l’aide de l’ensemble des resultats des 6 periodes reunies. Une valeur de Ir superieure ä 1 reflete donc une presence preferentielle de l’espece dans le milieu considere; une valeur de Ir proche de 1 signifie que l’espece est presente dans le milieu de fagon aleatoire; une valeur de Ir inferieure & 1 montre que la zone est plus ou moins delaissee par l’espece. Par ailleurs, les frequences de capture des especes dans les differentes zones de vegetation ont 304 L. Granjon et G. Cheylan permis de determiner deux parametres importants de la niche Ecologique de ces especes (BLONDEL 1979): - L’amplitude d’habitat AH = e!", avec H’ = -Zp;log,p; oü p; represente la frequence de l’espece dans le milieu ı. -Le barycentre G = x; +2x3+...+nx, / Zx, avec Xj, X2, X, representant les abondances des especes dans les milieux (n = 5 icı). Ce parametre situe le preferendum £Ecologique des especes dans le gradient considere. Analyse des interactions interspecifiques en laboratoire Les rythmes d’activite de souris et de mulots provenant de differentes localites, la plupart insulaires, ont Ete enregistres en absence et en presence de congen£res et de rats noirs. Pour ce faire, les anımaux sont places dans des cages individuelles reliees a un terrarium contenant nourriture et boisson ad libitum par l’intermediaire de tubes (fig. 2). Sur ces tubes, sont places des cellules photo-electriques reliees a un micro-ordinateur (REQUIRAND et al. 1987) enregistrant les passages des Rongeurs, ainsi que le temps qu’ils passent dans chacun des compartiments. L’indice d’activit€ pris en compte est le nombre de passage des cages individuelles au terrarium. Les resultats representent des moyennes de trois A six Jours d’exp£rience. Trois types d’enregistrement ont et& effectues: - rythme d’une souris ou d’un mulot seul dans le dispositif. -rythme de souris ou de mulot en presence d’un individu de la m&me esp£ce. - rythme d’une souris ou d’un mulot en presence d’un rat noir. Dans cette derniere configuration, le tube permettant le passage de la souris ou du mulot de sa cage au terrarıum a un diametre suffisamment petit pour que le rat noir ne l’emprunte pas. Fig. 2. Schema du protocole experimental d’enregistrement des rythmes d’activit€ de M. musculus et A. sylvaticus en presence ou absence de R. rattus ou d’un congenere; 1 = terrarium avec nourriture et boisson; 2= compteurs de passage; 3 = cages individuelles; 4 = interface; 5 = micro-ordinateur enregistrant les donnees Resultats Repartition altitudinale 131 souris, 129 rats, 21 mulots et 4 lerots (Eliomys quercinus) ont Ete captures au cours des 4124 nuits-pieges (n.p.) realısees dans l’ensemble de la Corse. Deux especes montrent une repartition semblable: R. rattus et M. muscnlus domesticus (Fig. 3). Ces deux especes sont particulitrement abondantes dans les zones humides littorales, ou leurs frequences sont comparables (0,058 et 0,054 ind/n.p.). Ces chiffres diminuent rapıdement dans les maquıis de l’etage collineen, ou le rat est pres de quatre foıs plus abondant que la souris, puis ces deux especes disparaissent progressivement vers 1000 m d’altitude. La repartition de A. sylvaticus est inversee par rapport ä celle des deux autres Murides: le mulot est generale- ment rare dans l’ensemble des &tages mediterraneen et supra-mediterraneen, oU ses abondances varient peu. Il devient relativement frequent (0,015 ind/n.p.) dans l’etage subalpın, oü les deux esp&ces precedentes sont absentes. Ces resultats correspondent etroitement avec ceux obtenus, dans l’ile, par Lisors (1984), d’apres l’examen de pelotes de la chouette effraie (T’yto alba): la frequence du mulot est en effet trois foıs plus faible dans les pelotes recueillies en-dessous de 500 m qu’en dessus de cette altitude, alors que l’inverse est vrai pour les deux autres Murides. En Provence continentale, ’abondance du mulot ALTITUDES 2100 1700 1500 1100 1000 900 Mecanismes de coexistence dans une gnilde de murides insulaires 305 ALTITUDES 2300 PELOUSES ET AULNE GLUTINEUX HETRAIE SAPINIERE ET PIN LARICIO 1200 CHATAIGNERAIE 005 700 RATTUS RATTUS MUS MUSCULUS APODEMUS APODEMUS SYLVATICUS SYLVATICUS CONTINENT CORSE Fıg. 3. Repartition altitudinale de R. rattus, M. musculus et A. sylvaticus en Corse, ainsi que d’A. sylvaticus sur le continent (d’apres Orsını 1981). Les frequences sont exprimees en nombre de captures par nuit/piege decroit regulierement avec l’altitude: la densite de l’espece est maximale dans les milieux mediterran&ens (0,068 ind/n.p.) et minimale dans l’etage subalpin (0,002 ind/n.p., Orsını 1981, cf fig. 3). La r£partition altitudinale de cette espece est donc completement differente en Corse et sur le continent. Or, dans les milieux mediterrandens continentaux, R. rattus est generalement rare ou absent et n’est abondant que dans quelques secteurs chauds. On peut donc supposer que la distribution du mulot en Corse est influencee par celle du rat noir, dont la r&partition est certainement dans cette ile plus reguliere que sur le continent, et les densites plus elevees. Densites et variations d’abondance Au cours des 6 sessions de piegeage r&alisees sur le quadrat d’Elbo, 138 captures de rats, 175 de mulots et 338 de souris ont &t& enregistrees, soit respectivement 88, 84, et 140 individus differents de chaque espece. Les densites des 3 Murides estim&es par l’indice de Lincoln ä partir de ces effectifs captures sont representes sur la figure 4. Le cycle de la souris et celui du rat sont correles positivement (r = 0,787 0,10 < p <0,05), alors que celuıi du mulot est correl& negativement avec celui des deux autres especes (avec souris r = -0,236; avec rat r= -0,309). Les chitfres obtenus varıent, pour le mulot entre 1 et 12 individus par hectare, pour le rat noir entre 3 et 14 ind/ha et pour la souris entre 3 et 23 ind/ ha. Par comparaison avec des donnees continentales obtenues dans les environs de Montpellier (Sud de la France) avec le m&me protocole, deux especes apparaissent nettement plus abondantes en Corse que sur le continent (M. musculus et R. rattus), alors que les densites de l’autre (A. sylvaticus) y sont plus faibles: dans un milieu dunaire, les densites de souris varıent de 0,5 & 7 ind/ha, alors que les densites de mulot fluctuent entre 9 et 28 ind/ha dans un milieu de garrigue (Orsını 1982). En Camargue, les densites de rat noir ne depassent pas 3 ind/ha (donnees non publiees). .068 306 L. Granjon et G. Cheylan individus/),, =, M. musculus 20 “A. sylvaticus AU R.rattus jan avr juil oct jan avr juil 1 9 8 4 — 1 9 8 5 Fig. 4. Variations d’abondance de R. rattus, M. musculus et A. sylvaticus de Janvier 1984 A juillet 1985 sur le quadrat d’Elbo (Corse). Les effectifs sont estimes par l’indice de Lincoln Ainsi, les effectifs de rats et de souris sont-ıls maxımums en fin d’automne-debut d’hiver, alors que ceux du mulot atteignent leur apoge&e au printemps. Pour cette derniere espece, le cycle demographique realıse en Corse est tres different de celui observe generalement sur le continent, en region me£diterraneenne: 3 populations de mulots, suivies entre 1979 et 1981 dans troıs biotopes continentaux differents des environs de Montpellier (France), montrent des pics de densite essentiellement hivernaux (Orsını 1982). Dans ces trois situations, les souris (Mus musculus domesticus ou Mus spretus ou les deux) sont les seuls autres Murides presents en effectifs comparables, le rat noir y Etant absent. De m&me, en Camargue, en absence de competiteur, les densites maximales de mulots sont nettement hivernales (JAmon 1986). A l’oppose, les varıations annuelles d’effectifs observees chez la souris a Elbo sont du me&me type que dans les populations continentales de l’espece (Orsını 1982; BOITANI et al. 1985; Navajas 1986). Les donnees concernant le rat noir sur le continent montrent une grande varıabilite des cycles de densite realıses (Davıs 1953). Dans d’autres iles mediterra- neennes, les densites de cette espece semblent en general assez fortes en hiver, en presence ou non d’autres Murides; la saison de reproduction est de plus beaucoup plus &talee en Corse que dans les petites iles (CHEYLAn 1986), ce qui permet au rat noir de voir sa population se reconstituer plus rapıdement que celle des deux autres especes a Elbo (cf. Fig. 4), apres le tres severe hıver 1984/1985. Repartition spatiale La figure 5 montre que la discrimination spatiale des trois especes est globalement tres nette, quelle que soit la periode consideree, a part en janvier 1984 (periode 1) ou rat noir et mulot ont des distributions semblables. Dans le detail, et au vu des histogrammes de la fig. 6, ıl apparait que: a. le rat noir est, A toutes les periodes A l’exception de Janvier 1984 (Ir = 0,94), capture preferentiellement dans la zone E, zone de recouvrement vegetal maximum (Ir = 1,84 a 2,10). Ceci se retrouve donc dans le schema global, prenant en compte toutes les sessions de capture (Ir = 1,53). Les autres milieux sont plus ou moins delaisses, a part le mılieu C, frequente de facon irreguliere selon les saisons. Mecanismes de coexistence dans une guilde de murides insulaires 307 Fıg. 5. Plan 1x 2 de ’AFC realısee a partir des nombres de captures de R. rattus, M. musculus et A. sylvaticus dans les 5 zones de vegetation (A A E) du quadrat d’Elbo (cf Fig. 1), au cours des 6 sessions de piegeage (1 & 6) b. la souris domestique est systematiquement ou presque sur-representee dans les milieux A (1,21 < Ir <2,02) et B (0,74 < Ir < 1,58), les deux milieux les plus ouverts. Elle est pratiquement absente de la zone C, tres rocheuse et se retrouve dans les zones D et E mais ä des effectifs assez faibles (Ir = 0,50 et 0,74 respectivement, sur l’ensemble des sessions). c. le mulot apparait globalement bien represente dans tous les milieux (0,73 < Ir <1,36 pour les cing zones, toutes sessions confondues), mais ceci traduit des abondances tres varıables dans chaque zone selon la periode de capture. Ainsi, il est recontre preferentielle- 308 L. Granjon et G. Cheylan R. rattus A.sylvaticus M. musculus Bi Milieux ä recouvrement vegetal creissant [cf figl ) c D E Fıg. 6. Indices de repartition (Ir, cf texte) par zone de vegetation, de R. rattus, M. musculus et A. sylvaticus sur le quadrat d’Elbo (Corse) Mecanismes de coexistence dans une guilde de murides insulaires 309 ment dans la zone C en janvier et juillet 1984, dans la zone D en octobre 84 et dans la zoneB en juillet 85. Il parait donc tres ubiquiste, avec un leger preferendum pour les milieux ä recouvrement vegetal intermedıaire. Le rat noir apparait finalement comme l’espece la plus forestiere, avec un barycentre (G) de 3,6 (Tab. 1), alors que la souris est caracteristique des milieux ouverts (G = 2,3). Le mulot est l’espece qui presente l’amplitude d’habitat la plus grande (AH = 4,2 contre 3,5 et 3,1 aux deux autres especes). Tableau 1. Amplitude (AH) et barycentre (G) d’habitat de R. rattus, M. musculus et A. syl- vaticus sur le quadrat d’Elbo R. rattus A. sylvaticus M. musculus AH 3,48 4,16 3,14 G 3,58 2,35 Dt Rythmes d’activite en captivite Dans tous les cas, le rythme d’activite des rats noirs testes s’est montr& independant de la presence ou non d’un individu d’une autre espece. Il n’est pas non plus apparu de differences entre les rythmes d’activite des souris et des mulots seuls ou avec un consp£cifi- que. Par contre, le comportement des souris (N = 4) et mulots (N = 4) a Et€ diversement affecte par la presence d’un rat. En ce qui concerne la souris (Fig. 7), on observe generalement une reduction de l’activite d’un facteur 2 en presence du rat noir, et/ou, dans deux cas, un decalage de l’activite vers le matin. Ces differences sont dans deux cas significatives (test de Wilcoxon, SIEGEL 1956). Dans un cas la souris a Ete finalement tu&e par le rat, le troisıeme jour de l’experience. Les rythmes d’activite des mulots testes ont EtE tres perturbes par la presence du rat noir (Fig. 8): dans deux cas, on observe une diminution d’un facteur 2 et dans un cas une augmentation d’un facteur 7 de la quantite d’activite. Ces differences sont toutes significa- tives. Dans 3 cas sur 4, le mulot a Ete tue et partiellement devore par le rat au bout de 3 ou 4 jours d’experience. Discussion Plusieurs auteurs ont insiste sur l’appauvrissement sp£cifique qui caracterise les peuple- ments ınsulaires (MAc ARTHUR et WıLson 1967; BLONDEL 1979, 1986; WILLIAMSON 1981; Brown et GiBson 1983). Cet appauvrissement est souvent compense par l’augmentation de densite des especes presentes, generalement bien plus abondantes dans les iles que sur le continent. Ainsi, P’appauvrissement specifigque du peuplement mammalıen de la Corse est impor- tant: par rapport A des zones homologues du continent, 62 % des especes manquent en Corse, la communaute de Rongeurs ne comptant que 6 especes (CHEYLAN 1984); celles-ci se repartissent en 4 Murides (Rattus norvegicus plus les 3 especes Etudiees ıcı) et 2 Glirides (Glis glis et Ehomys quercinus). Toutefois, les repartitions de 2 de ces especes (R. norvegicus et G. glis) sont tres ponctuelles, alors que les autres sont assez bien reparties dans la plupart des milieux de l’ile (SaLoTTı 1984). On peut donc s’attendre A trouver dans cette le une compensation de densite affectant ces 4 especes. Effectivement, on constate que les abondances du rat et de la souris sont plus &levees en Corse que dans des biotopes sımilaires du Midi de la France et d’Italie, alors que les 310 L. Granjon et G. Cheylan [N] 10 | ae - 48 20 15 10 * 5 > -10 10 KrkOo% = => 29 12 S N - 18 > ; -n 12 24 2 Rat absent Rat present ) zZ un Fig. 7. Rythmes d’activite de souris (N = 4) en absence et en presence de rat noir (abscisse: heures; ordonnee = nombre de passages. Test de Wilcoxon: NS = non significatif; *p<0,05; ***p <0,001) Mecanismes de coexistence dans une guilde de murides insulaires 311 10 NS - N-67 5 KK E ; > N-14 24 12 24 12 Rat absent Rat present Fıg. 8. Rythmes d’activite de mulots (N = 4) en absence et en presence de rat noir (m&me legende que pour la souris) abondances du mulot sont comparables (Orsını 1982; CassaınG et CRoSET 1985; BOTTANI et al. 1985; Navajas 1986; CHEYLAN 1986). Ces augmentations de densites insulaires sont generalement attrıbuees ä une diminution des pressions de predation et de competition intersp£cifique (LIDICKER 1973; TAMARIN 1977; CHEYLAN et GRANJON 1985). En l’absence d’augmentation de densite chez A. sylvaticns, on peut penser que les pressions de competition intersp£cifique vis-a-vis de cette esp£ce ne se sont pas relächees dans le secteur BD L. Granjon et G. Cheylan d’etude, la predation pouvant £tre par ailleurs consideree comme s’exercant de la m&me facon sur les trois especes, ou tout au moins sur le mulot et la souris, de taille comparable. Le decalage temporel du cycle demographique du mulot par rapport ä celui des deux autres Murides est du m&me type que celui recontre en Italie entre M. musculus domesticus et A. sylvaticus par BOITANI et al. (1985), dans un milieu littoral ou vit Egalement R. rattus (BOITANI, comm. pers.). Pour ces auteurs, ce me&canisme aurait pour but de reduire la competition entre les deux especes et aurait donc un caractere adaptatıf. Dans notre cas, le meme phenomene peut Etre invoque, pour permetre la coexistence @quilibree des trois especes. Le rat noir doit £tre le responsable du decalage du cycle demographique du mulot par rapport ä ce qu’il est habituellement dans les biotopes mediterraneens continentaux oü Rattus rattus est absent (Orsını 1982; JAmoN 1986). Un autre mecanisme qui semble entrer en jeu pour favoriser la coexistence entre les Murides est la segregation spatiale de l’habitat. En effet, les trois especes Etudiees se repartissent differemment dans le gradient de milieux defini (Fig. 5 et 6): Mus musculus est beaucoup plus abondante dans les milieux ouverts, ce qui confirme ce que l’on sait des habitudes Ecologiques de cette espece, liee le plus souvent aux milieux A faible recouvre- ment arbustif et arbore (Orsını 1982; ORrsını et al. 1982; BoITAnT et al. 1985). Rattus rattus est quant & luı plus frequemment capture dans les zones boisees qui constituent son habitat preferentiel, puisqu’en region mediterran&enne continentale, il est infeode & la foret ou ıl est souvent arboricole (SAINT-GIRONS 1973). Le mulot est generalement tres ubiquiste, mais montre luı aussi une preference pour les milieux assez forestiers (CORBET 1966; NIETHAMMER et KrapPp 1978). En region mediterraneenne continentale, dans un biotope heterogene et en presence de Mus spretus, ıl apparait eEgalement lie A un couvert vegetal important: garrıgue A chene kermes (Q. coccifera) et taillis A chene vert (Q. ılex) (Orsını 1982). En revanche, ıl est assez bien r&parti sur l’ensemble des milieux definis & Elbo, alors que son optimum d’habitat se rapprocherait de celui de R. rattus. Il semble donc que A. sylvaticus abandonne & Elbo les milieux les plus fermes pour se retrouver dans les milieux intermediaires et m&me ouverts. Celä est possible gräce A ses capacites ä vivre dans des habitats tres dıvers ce que confirme ici sa grande amplitude d’habitat (Tab. 1), qui n’est cependant pas superieure en Corse ä ce qu’elle est sur le continent, contrairement aux deux autres Murides (LıBo1s 1984). C’est lA encore le rat noir qui semble & l’origine de la repartition observee du mulot dans la zone d’etude. La differenciation temporelle (decalage des cycles demographiques) et spatiale (segrega- tion ditferentielle de l’habitat) entre les troıs Murides du vallon d’Elbo, en permet sans doute une exploitation mieux Equilibree des diverses ressources. Cette differenciation de niches est attestee par les nombres assez reduits de captures de deux individus d’especes differentes au m&me jalon la m&me nuit, par rapport aux chiffres attendus sous l’hypothese d’une repartition aleatoıre des individus sur l’ensemble du quadrat ä chaque session de piegeage (tab. 2): l’evitement mulot-rat est maximum alors que ce sont les deux esp£ces les plus susceptibles de se rencontrer, d’apres leur repartition sur le quadrat (ct. Fig. 5 et 6). Tableau 2. Nombres theoretique et observe de captures au me&me jalon et la m&me nuit de 2 Murides d’especes differentes sur le quadrat d’Elbo (X? = 27,7; p 0,001) Rat Mulot Rat Souris Mulot Souris Nombre Theorique 26 33 Nombre Observe 13 13 Le me&canisme ä l’origine de la segregation spatio-temporelle des trois especes pourrait etre de nature comportementale. L’influence de la presence du rat noir sur l’activite des deux autres Murides montre que ces dernieres especes sont amenees a developper en Mecanismes de coexistence dans une guilde de murides insulaires 313 captivite des me&canısmes d’evitement (deplacement ou reduction de l’activite) pour pou- voir cohabiter avec le rat noır. Il apparait m&me dans ces conditions qu’un rat noir est capable du tuer assez rapıdement un mulot ou une souris partageant son espace vital, le mulot semblant particulierement devoir souffrir d’une telle promiscuite ce qui confirme les faits observes dans le milieu naturel, ou le decalage des cyeles demographiques permet cette coexistence, de m&me que la possibilite d’occuper des micro-habitats differents. TERMAN (1974) montre par exemple que la cohabitation de Sigmodon hispidus et Microtus ochro- gaster en captivite est possible seulement sı le milieu est complexifi€ par adjonction de fragments de papiers. Dans ce cas toutefois, les deplacements de M. ochrogaster sont beaucoup reduits par la presence de $. hispidus. De m&me, les deplacements du mulot en Corse sont particulierement faibles par rapport A ce qui a &t€ observe dans un milieu continental similaire, et avec le m&me protocole: (Distance Maximale de Recapture = 16 m a Elbo, 37 m vers Montpellier, Sud de la France, Orsını 1982). A la suite d’une etude de plusieurs guildes de micromammiferes deserticoles syntopi- ques, Harrer (1982) et Harrer et al. (1983) proposent un me&canisme selon lequel les especes les plus gen£ralistes seraient Egalement les moins bonnes competitrices et verraient leur amplitude d’habitat se reduire lors de l’augmentation de densite des especes plus specialisees et meilleures competitrices, hypothese tout A fait coherente avec le «taxon cycle» de Mac ARTHUR et Wırson (1967). Dans le cas de la guilde des Murid&s corses, A. sylvaticus, l’espece la plus generaliste, parait effectivement Etre le «moins bon» competiteur des trois especes en presence: ıl est domine nettement par le rat noir, et les exp£riences de confrontation avec A. flavicollis (HOFFMEYER 1973; MONTGOMERY 1978) montrent qu’il peut Egalement l’Etre par une espece de taille comparable & la sienne. A l’ıssue de cette approche des relations interpopulationnelles et interindividuelles ä differentes Echelles de perception, on peut tenter de preciser le schema connu de la colonisation puis de l’etablissement d’Apodemus sylvaticus, Rattus rattus et Mus musculus enu@orse: Au cours du Pleistocene et du tout debut de l’Holocene, la communaute de Rongeurs de la Corse n’est composee que de deux especes end&miques: Rhagamys orthodon (Muride) et Thyrrenicola henseli (Microtide) (ViGnE 1983a et b). Ces deux esp£ces persistent jusqu’ä la fin du premier Millenaire BC et coexistent donc pendant 3000 ans avec A. sylvaticus, qui apparait dans l’ile au debut du 3°”° Millenaire BC. En revanche, R. rattus n’est signal& qu’au 6°”° siecle apres J.C. (VIGNE et MARINGAL- VIGnE 1985), alors qu’aucun reste fossile de Mus n’est connu en Corse. Neanmoins, ces deux especes pourraient avoir colonise l’ile plus töt que ne l’indiquent les donnees paleontologiques actuellement disponibles, car elles sont signalees entre le Neolithique et l’epoque romaine dans plusieurs iles de Mediterranee (Minorque: REUMER et SANDERS 1984; Sardaigne: SAnGES et ALCOVER 1980; Malte: STORCH 1970). Etant donne la taille de Rhagamys et Thyrrenicola, que l’on peut estimer ä environ 50 g (MıcHAux, comm. pers.), ıl est probable que ces especes pouvaient dominer Apodemus en condition de syntopie. Les restes fossıles de ces deux Rongeurs sont d’ailleurs abondants dans les gisements jusqu’ä la fin du 1° millenaire, alors qu’Apodemus n’a livre que peu de restes osseux de cette &poque. La colonisation de la Corse par cette esp£ce aurait donc pu etre assez lente. Par contre, la disparition rapide et sımultanee des deux especes endemiques est plus ou moins concomitante ä l’expansion geographique de R. rattus en Europe (ARMITAGE et al. 1984). Parmi les differentes causes de l’extincetion de Rhagamys et Thyrrenicola revues par VIGNE (1983), ’apparition dans l’ile de R. rattus pourrait bien £tre la principale. Au cours de ses 5000 ans de presence dans l’ile, Apodemus sylvaticus a donc toujours ete domine par des especes de taille superieure ä la sienne, especes qui exergaient, ou qui exercent, une pression de competition importante sur ses populations. Neanmoins, pour des raisons liees ä leurs origines biog&ographiques, ıl est peu probable 314 L. Granjon et G. Cheylan que M. musculus et R. rattus, especes thermophiles, puissent coloniser A court terme les biotopes forestiers montagnards a climat froid habites par A. sylvaticus, E. quercinus, et dans une moindre mesure, G. glıs. A. sylvaticus est donc la seule espece, dans la guilde des Murides corses, pouvant habiter les milieux forestiers de moyenne altitude caracterises par un relächement de la competi- tion. Elle represente donc, dans la theorie du «taxon cycle», le meilleur candidat A une differenciation insulaire, ce qui a Et& verifi€ par une &tude biometrique portant sur 26 caracteres cränıens et corporels, comparant les populations de 4 Rongeurs de la Corse et du Midi de la France: A. sylvaticus, M. musculus, R. rattus et E. quercinus (ORsINI et CHEYLAN 1983). Cette etude a revele que la differenciation morphologique la plus poussee etait atteinte par Apodemus, chez qui tous les caracteres Etudies sont plus grands en Corse que sur le continent, alors que les trois autres especes &tudiees ne sont en general differenciees que sur des caracteres de l’appareil manducateur. Lisoıs et al. (1983), dans une comparaison entre les mulots de Corse et de la Belgique obtiennent des resultats comparables. Cecı est d’autant plus significatif que le groupe A. sylvaticns/A. flavicollis en Europe sud-occidentale se caracterise par une grande stabilite morphologique au cours des 3 derniers millions d’annees (MıcHAux 1983). Ces r&sultats corroborent donc les predictions du modele de Mac ARTHUR et WILSON (1967). La repartition tres etendue mais fractionnee d’Apodemus sylvaticus dans Pile entraine sans doute une reduction des flux geniques, alors que les populations de Rattus rattus et, dans une moindre mesure, de Mus musculus, limitees A la ceinture de milieux mediterraneens qui entoure les montagnes de l’ile, sont soumises ä des Echanges geniques plus intenses qui ralentissent leur ditferenciation. Remerciements Ces recherches ont &t& partiellement financees par le Parc Naturel Regional de Corse, gräce A son Directeur, M. LEENHARDT, que nous tenons & remercier ıcı. Notre reconnaissance va egalement 3 a J. ©. THIBAULT, J. M. CasTa, J. M. CaıLLaup et C. H. Bıancont pour leur appui lors des missions dans l’ile, de me&me qu’& R. Fons, J. P. CLara, O. POULIQUEN, M. Navajas, J. C. Aurrray, A. CLAMENS et H. GAUBERT qui ont participe au travail de terrain. P. Orsını nous a confie ses notes de piegeage qui ont permis de pre£ciser la r&partition des especes dans l’ile. J. BLONDEL, J. Cassaınc et]. MICHAUX nous ont fait part de leurs remarques et nous les en remerciont vivement, ainsı que H. CROSET, initiateur des ces recherches. Resume La guilde des Murides de Corse, form&e du rat noir (Rattus rattus), du mulot sylvestre (Apodemus sylvaticus) et de la souris domestique (Mus musculus domesticus) est etudiee A trois nıveaux de perception differents: A l’echelle regionale, la repartition altitudinale du mulot est tres differente de celle qu’il a sur le continent en region mediterraneenne. Au niveau stationnel, une etude sur quadrat de piegeage montre que le cycle d’abondance du mulot est decale par rapport a celui des deux autres Murides et par rapport a ce qu’il est sur le continent. D’autre part, les repartitions spatiales des trois especes sont complementaires, chacune £tant preferentiellement rencontree dans un type donne d’habitat. Enfin, les r&sultats de tests en captivite font apparaitre une influence negative nette du rat noir sur l’activite des deux autres especes. Cet ensemble de donnees suggerent que la coexistence entre les troıs especes n’est possible que gräce ä un ajustement de leurs distributions temporelle et spatiale. Par ailleurs, le mulot qui a Et& le premier colonisateur de la Corse parmi ces trois especes, semble Etre actuellement l’espece qui s’est le plus differenciee. La competition avec les deux autres especes pourrait etre a l’origine de cette Evolution. Zusammenfassung Mechanismen der Koexistenz von Insel-Muriden (Rattus rattus L., Apodemus sylvaticus L. und Mus musculus domesticus Rutty) auf Korsika: Evolutionäre Konsequenzen Untersuchungen an den drei Muriden-Arten Rattus rattus, Apodemus sylvaticus und Mus musculus domesticus wurden auf Korsika (Frankreich) durchgeführt. Es ergaben sıch folgende Ergebnisse: Die Höhenverteilung in der Häufigkeit des Auftretens unterscheidet sich bei Apodemus sylvaticus auf Korsika stark von der auf dem mediterranen Festland. Auch die jahreszeitliche Abundanz-Änderung Mecanismes de coexistence dans une guilde de murides insulaires 315 dieser Art weicht von der auf dem Festland ab und darüber hinaus auch von jener der beiden anderen Muriden-Arten auf Korsika. Die drei Arten besiedeln bevorzugt Habitate, die einander ergänzen. Versuche in Gefangenschaft zeigten, daß Rattus rattus die Aktivitäten der beiden anderen Arten herabsetzt. Offenbar bestimmt die auf Korsika im Vergleich zum Festland häufigere Hausratte die dort andersartige Verteilung der Waldmaus. Apodemus sylvaticus, der erste Einwanderer unter diesen drei Arten, ist dieser interspezifischen Konkurrenz am stärksten und in besonderer Weise ausgesetzt. Möglicherweise i ist aus entsprechenden Gründen in erster Linie Rattus rattus verantwortlich für den Zusammenbruch von Populationen und das Verschwinden der endemischen Muriden, die bis 1000 v.Chr. auf Korsika gelebt haben. Bibliographie ARMITAGE, P.; West, B.; STEEDMAn, K. (1984): New evidence of Black Rat in roman London. London Archeologist 4, 375-383. Bronner, J. (1979): Biogeographie et Ecologie. Parıs: Masson Ed. BLonDeL, ]J. (1986): Biogeographie Evolutive. Parıs: Masson Ed. Bortanı L.; Loy, A.; MoLinart, P. (1985): Temporal and spatial displacement of two sympatric rodents (Apodemus sylvaticus and Mus musculus) in a mediterranean coastal habitat. Oikos 45, 246-252. BRADLEY, R. A.; BrapLey, D. W. (1985): Do non-random patterns of species in niche space imply competition? Oikos 45, 443445. Brown, J. H.; GıBson, A. C. (1983): Biogeography. C. V. Mosby Company. Carrauist, S. (1966): Loss of dispersability in Pacific Compositae. Evolution 20, 3048. Cassaıng, J.; Croset, H. (1985): Organisation spatiale, competition et dyramique des populations sauvages de souris (Mus spretus et Mus musculus domesticus) du Midi de la France. Z. Säugetier- kunde 50, 271-284. CHEYLAn, G. (1984): Les mammiferes des iles de Provence et de Mediterranee occidentale: un exemple de peuplement insulaire non Equilibre? Rev. Ecol. (Terre Vie) 39, 37-54. CHEYLAN, G. (1986): Facteurs historiques, €Ecologiques et genetiques de l’evolution des populations mediterraneennes de Rattus rattus (1..). These d’Etat, Univ. Montpellier. CHEYLAN, G.; GRANJON, L. (1985): Ecologie d’une population de rats noirs (Rattus rattus) A Port- Cros (Var). Methodologie et premiers resultats obtenus sur quadrat. Trav. Sci. Parc Nat. Port Cros, Fr. 11, 109-131. ConnELL, J. H. (1983): On the prevalence and relative importance of interspecific competition: evidence from field experiments. Am. Nat. 122, 661-696. CoRBET, G. B. (1966): The terrestrial mammals of Western Europe. Londres. CROWELL, K. L. (1973): Experimental zoogeography: introductions of mice to small islands. Am. Natur. 107, 535-559. CROWELL, K. L. (1983): Islands — Insight or artifact? Population dynamics and habitat utilization in insular rodents. Oikos 41, 443-454. CROWELL, K. L.; Pımm, $. L. (1976): Competition and niche shifts of mice introduced onto small ıslands. Oikos 27, 251-258. Dueser, R. D.; PoRTER, J. H. (1986): Habitat use by insular small mammals: relative effects of competition and habitat structure. Ecology 67, 195-201. GLiwıcz, J. (1980): Island populations of rodents: their organization and functionning. Biol. Rev. 55, NO ER HALLET, J. G. (1982): Habitat selection and the community matrix of a desert small mammal fauna. Ecology 63, 1400-1410. HALLET, ]J. G.; O’ConnELı, M. A.; HonEycUTT, R. L. (1983): Competition and habıtat selection: test of a theory using small mammals. Oikos 40, 175-181. HOoFrFMEYER, L. (1973): Interaction and habitat selection in the mice Apodemus flavicollis and A. sylvaticus. Oikos 24, 108-116. Jamon, M. (1986): The dynamics of Wood mouse (Apodemus sylvaticus) populations in the Camar- gue. J. Zool. Lond. 208, 569-582. Lisois, R. M. (1984): Le regime alimentaire de la chouette effraie. Cahiers d’Ethologie Appliquee 4, 1202. LıBo1s, R.; BORDENAVE, D.; Fons, R. (1983): Biometrie cränıenne du mulot de Corse: contribution A une etude taxonomique. Bull. Ecol. (sous presse). LiDICkErR, W. Z., Jr. (1973): Regulation of numbers in an island population of the California vole, a problem of community dynamics. Ecol. Monog. 43, 271-302. Mac ARTHUR, R. H. (1972): Geographical Ecology. New York: Harper and Row. Mac ARTHUR, R. H.; Wırson, E. ©. (1963): An equilibrium theory of insular zoogeography. Evolution 17, 373-387. Mac ARTHUR, R. H.; Wırson, E. ©. (1967): The theory of island biogeography. Princeton: Princeton Univ. Press. 316 L. Granjon et G. Cheylan MıcHAux, J. (1983): Aspects de l’evolution des Murides (Rodentia, Mammalia) en Europe Sud- Occidentale. Coll. Int. CNRS n°330, 195-199. MONTGOMERY, W. 1. (1978): Intra and interspecific interactions of Apodemus sylvaticus (L.) and A. flavicollis (Melchior) under laboratory conditions. Anım. Behav. 26, 1247-1254. Navajas, M. J. (1986): Facteurs selectifs et stochastiques de la differenciation de populations insulaires de Mus musculus domesticus. These 3°”“ cycle, Univ. Montpellier. NIETHAXER, J.; Krapp, F. (1978): Handbuch der Säugetiere Europas. Wiesbaden: Akademische Verlagsges. Orsını, P. (1981): Premiers Echantillonnages de micromammiferes en Provence. Mammalia 45, 187-197. Orsını, P. (1982): Facteurs regissant la repartition des souris en Europe: interet du modele souris pour une approche des processus evolutifs. These 3eme cycle. USTL Montpellier. Orsinı, P.; CAssaınG, J.; DUPLANTIER, J. M.; CrosET, H. (1982): Premieres donnees sur l’ecologie des populations naturelles de souris Mus spretus Lastate et Mus musculus domesticus Rutty dans le midi de la France. Rev. Ecol (Terre Vie) 36, 321-336. Orsını, P.; CHEyYLAn, G. (1983): Les Rongeurs de Corse: modifications de taille en relation avec l’ısolement. Bull. Ecol. (sous presse). PacaLa, S. W.; ROUGHGARDEN, J. (1985): Population experiments with the Anolıs lizards of StMaarten and St Eustatius. Ecology 66, 129-141. REQUIRAND, C.; POULIQUEN, O.; GRANJON, L.; CROSET, H. (1987): Description et utilisation d’un systeme automatise de compteurs de passages pour micro-mammiferes. Mammalia (sous presse). REUMER, J. W. F.; SAnDERS, E. A. C. (1984): Changes in the vertebrate fauna of Menorca in prehistoric and classıcal times. Z. Säugetierkunde 49, 321-325. RıckLers, R. E. (1970): Stage of taxon cycle and distribution of birds on Jamaica; Greater Antilles. Evolution 24, 475-477. RıckLers, R. E.; Cox, G. W. (1972): Taxon cycle in the West Indian avifauna. Am. Nat. 106, 195-219. RıckLErs, R. E.; Cox, G. W. (1978): Stage of taxon cycle, habitat distribution, and population density in the avıfauna of the West Indies. Am. Nat. 112, 875-895. RumMEL, J. D.; ROUGHGARDEN, J. (1985): Effects of reduced perch-height separation on competition between two Anolıs lizards. Ecology 66, 430-444. SAaInT GIRoNS, M. C. (1973): Les mammiferes de France et du Benelux. Parıs: Doin Ed. SarorTı, M. (1984): Atlas regional des mammiferes de la Corse. Ajaccıo. SANGES, M.; ALCOVER, J. A. (1980): Noticia sobra la microfauna vertebrada holocenia de la grotta Su Guanu o Gonagosula (Olıena, Sardenya). Endins 7, 57-62. SCHOENER, T. W. (1983): Field experiments on interspecific competition. Am. Nat. 122, 240-285. SIEGEL, $. (1956): Nonparamertric Statistics for the behavioral sciences. Maidenhead: McGraw-Hill. STORCH, G. (1970): Holozäne Kleinsäugerfunde aus der Ghar Dalam-Höhle, Malta. Senckenbergiana Biologica 51, 135-145. TAMARIN, R. H. (1977): Dispersal in ısland and mainland voles. Ecology 58, 1044-1054. TERMAN, M. R. (1974): Behavioral interactions between Microtus and Sigmodon: A model for competitive exclusion. J. Mammalogy 55, 705-719. VIGNE, J. D. (1983a): Le remplacement des faunes de petits mammiferes en Corse lors de l’arrıvee de ’homme. C. R. Soc. Biogeog 59, 41-51. | VIGNE, J. D. (1983b): Les mammiferes terrestres non-volants du Post-glaciaire de la Corse. These 3°” cycle, Univ. Parıs VI. VIGNE, J. D.; MARINGAL-VIGNE, M. C. (1985): Le rat en Corse au 6°” sıecle? Mammalıa 49, 138-139. WILLıamsoN, M. (1981): Island populations. Oxford: Oxford Univ. Press. Wırson, E. ©. (1961): The nature of the taxon cycle in the Melanesian ant fauna. Am. Nat. 95, 169-193. Adresse des auteurs: LAURENT GRANJON et GILLES CHEYLAN, Laboratoire d’Eco-Ethologie, Institut des Sciences de l’Evolution, Unite Associee au CNRS n° 327, Universite des Sciences et Techniques du Languedoc, Place Eugene Bataillon, F-34060 Mont- pellier Cedex, France Z. Säugetierkunde 53 (1988) 317-319 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 DIISSENSEELIETISTERIETRURZNIEITEIUNE Weitgehende Rotation des 4. Prämolaren im Unterkiefer eines Rothirsches (Cervus elaphusL.) und eines Rehbockes (Capreolus capreolus L.) Von U. KıERDORF und H. KIERDORF Zoologisches Institut der Universität zu Köln Eingang des Ms. 01. 02. 1988 Als Zahnrotation (Rotatio dentis) bezeichnet man eine Stellungsanomalie, bei der ein Zahn um seine Längsachse gedreht ıst. Im folgenden werden zwei Fälle sehr weitgehender Rotation des mandibularen 4. Prämolaren bei Cerviden beschrieben. Bei dem ersten Objekt handelt es sich um die linke Unterkieferhälfte eines Rothirsches (Cervus elaphus), dessen Alter nach dem Zahnabschliff auf 6-7 Jahre geschätzt wird. In der mit 11,96 cm (Messung an Kronenbasis) normal langen Backenzahnreihe (BÜTZLER 1986) ist der P, um 180° rotiert, so daß seine morphologische Buccalseite nach lingual weist (Abb. 1). Die (typisch geformte) Krone des Zahnes ıst gegen dıe Wurzel buccalwärts abgebogen (Buccaldeviation der Zahnkrone) und schert daher leicht aus der Backenzahn- reihe aus. Fine Torsion, d. h. eine Verdrehung der Wurzel gegenüber der Zahnkrone, tritt jedoch nicht auf. Der rotierte P, kann nicht aus seiner Alveole herausgezogen werden, da die Wurzelspitzen, wie Röntgenaufnahmen zeigen, mesialwärts umgebogen sind. Die mesiale und distale Kronenfläche dieses Zahnes sowie die Approximalflächen der angren- zenden P; und M, zeigen deutliche Abnutzungsspuren. Als zweites Objekt liegt die rechte Unterkiefer-Backenzahnreihe eines etwa zweijähri- gen Rehbockes (Capreolus capreolus) vor, deren Länge mit 6,16 cm (Messung an Kronen- basıs) innerhalb der für diese Art festgestellten Variationsbreite liegt (LEHMANN und SÄGESSER 1986, eigene Befunde). In diesem Fall beträgt die Rotation des P, etwa 140° (Abb. 2), so daß die Distalfläche des Zahnes mesiobuccalwärts weist. Zwischen ıhm und den benachbarten Zähnen haben sich infolge approximaler Abnutzung distinkte Kontakt- flächen ausgebildet. Die Krone des gedrehten Zahnes ist normal gestaltet. Das Auftreten von Rotatio dentis ıst in der odontologischen Literatur vielfach belegt (EIDMANN 1939; GARLICK 1954; DE JONGE 1965; BECKER 1970; PınDBoRG 1970; MEYER 1975; VıGaL und MACHORDOM 1987). In der Regel beträgt die Drehung maximal 90° und tritt zumeist als Folge einer zu engen Zahnstellung auf. Auch durch Hypodontie oder posteruptiven Zahnverlust ausgelöste Migration einzelner Zähne kann zur Entstehung derartiger Stellungsanomalıen führen. Zahnrotationen von mehr als 90° in vollständigen Zahnreihen sind demgegenüber sehr selten. Im menschlichen Gebiß zeigt nach DE JonGe (1965) am häufigsten der P? eine solche Veränderung, die gelegentlich auch bilateral-symmetrisch auftreten kann. Für Cerviden liegen nur wenige Mitteilungen über weitgehende Zahnrotationen vor. EıpmAnn (1939) schildert die Drehung des linken I, um 180° bei einem Rothirsch. Bei der gleichen Art beschreiben PucHEr (1983) und WOLLENHAUPT (1986) jeweils eine 180° betragende Rotation des linken P,. Belegt wird die sehr geringe Frequenz des Auftretens dieser Aberration durch die Tatsache, daß bei der von den Verfassern im Rahmen einer U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5305-0317 $ 02.50/0 318 U. Kierdorf und H. Kierdorf Abb. 1. Um 180° gedrehter linker P, (*) eines Rothirsches. Ansicht von okklusal. Beachte Abnutzung der approximalen Zahnflächen (Pfeil) - Abb. 2. Um ca. 140° rotierter P, (*) in der rechten Unterkiefer- Backenzahnreihe eines Rehbockes. Ansicht von bucco-okklusal anderen Untersuchung (KIERDORF und KIERDORF 1986) durchgeführten Überprüfung von ca. 10 000 Reh-Unterkiefern einzig der hier dargestellte Fall beobachtet wurde. Die Ursachen der über 90° hinausgehenden Zahnrotationen sind zur Zeit noch nicht geklärt. DE JonGE (1965) stellt sie mit Recht als Anomalien besonderer Art heraus, die nicht als Folge von Raummangel im Zahnbogen erklärt werden können. In diesem Zusammenhang ist auf die Untersuchungen von Oo& (1962, 1968) am menschlichen Gebiß und von EsakA (1982) an demjenigen dolichocephaler Hunde hinzuweisen. Danach führen die Zahnkeime ım Normalverlauf der Odontogenese bereits auf dem Knospen- und Kappenstadıum Rotationsbewegungen aus. Dies geschieht während einer Entwicklungs- phase, in der zwischen den einzelnen Anlagen noch ausreichende Zwischenräume vorhan- den sind. Die Rotationen lassen sich daher nach Ansicht der genannten Autoren nicht auf Platzmangel zurückführen, sondern sind Folge autonomer Bewegungstendenzen der Zahnkeime. Im Zuge der Normalentwicklung werden die Drehungen gegenüber der definitiven Position der Zähne im Kiefer später rückgängig gemacht. Vermutlich liegt den beschriebenen Zahnstellungsanomalien eine Störung der oben geschilderten Vorgänge zugrunde. Rotation des 4. Prämolaren im Unterkiefer von Rothirsch und Reh 319 Danksagung Für die Überlassung des Materials danken die Verfasser Herrn L. v. Lünınck, Bourheim, und Herrn Forstamtmann a. D. Marx, Rheurdt. Literatur BECKER, E. (1970): Zähne. In: Handbuch der speziellen pathologischen Anatomie der Haustiere. 3. Aufl. Bd. 5, Teil 1. Hrsg. von E. Jost. Berlin, Hamburg: Paul Parey. 83-313. BÜTZLER, W. (1986): Cervus elaphus Linnaeus, 1758 — Rothirsch. In: Handbuch der Säugetiere Europas. Bd. 2/II. Hrsg. von J. NIETHAMMER und F. Krapp. Wiesbaden: Aula. 107-139. DE Jonge, T. E. (1965): Rotatio dentis. Morph. Jb. 108, 67-70. Eıpmann, H. (1939): Untersuchungen am Gebifß des Rothirsches und der anderen einheimischen Cerviden. Hannover: M. & H. Schaper. Esara, S. (1982): Development of rotation of mandibular premolar tooth germs in the dog. Acta anat. 114, 211-227. GARLICK, N. L. (1954): The teeth of the ox in clinical diagnosis III. Developmental anomalies and general pathology. Am. J. Vet. Res. 15, 500-508. KIERDORF, U.; KIERDORF, H. (1986): Erste Untersuchungsergebnisse zum Auftreten von Dentalfluo- rose beim Rehwild (Capreolus capreolus L.) in Nordrhein-Westfalen. Z. Jagdwiss. 32, 55-56. LEHMANN, E. von; SÄGESSER, H. (1986): Capreolus capreolus Linnaeus, 1758 — Reh. In: Handbuch der Säugetiere Europas. Bd. 2/lI. Hrsg. von J. NIETHAMMER und F. Krapp. Wiesbaden: Aula. 233-268. Meyer, P. (1975): Beispiele angeborener Zahn- und Gebißanomalien beim Europäischen Reh (Capreolus capreolus Linne, 1758) nebst einigen Bemerkungen zu deren Genese und Terminologie. Z. Jagdwiss. 21, 89-105. Oo#, T. (1962): On the development of position of the human deciduous molar germs in the latter half of the fetal life. Okajımas Fol. anat. jap. 38, 51-71. — Changes of position of deciduous molar and premolar germs during development. Okajımas Fol. anat. jap. 44, 83-97. PınDBorg, J. J. (1970): Pathology of the dental hard tissues. Copenhagen: Munksgaard. PucHER, E. (1983): Eine seltene Zahnanomalıe an einer subfossilen Mandibel von Cervus elaphus L. Z. Jagdwiss. 29, 248-250. VıGAL, C. R.; MACHORDOM, A. (1987): Dental and skull anomalies in the Spanish wild goat, Capra pyrenaica Schinz, 1838. Z. Säugetierkunde 52, 38-50. WOLLENHAUPT, H. (1986): Zahnanomalie beim Rothirsch. Jäger 104, H. 8, 8. Anschrift der Verfasser: UwE KIERDORF und HorsT KIERDORF, Zoologisches Institut der Universität zu Köln, I. Lehrstuhl, Weyertal 119, D-5000 Köln 41 320 BUEHBESERESFIINISEIN MossMmann, H. W.: Vertebrate Fetal Membranes. Comparative Ontogeny and Morphol- ogy, Evolution, Phylogenetic Significance, Basic Functions, Research Opportunities. Basıngstoke: McMillan Press, and New Brunswick, N. J.: Rutgers University Press 1987. 383 pp. £ 80.00; US $ 95.00. ISBN 0-8135-1132-1 Der Autor dieses Buches publizierte 1937 sein heute klassisches Werk „Comparative Morphogenesis of Fetal Membranes“. Das neue Buch von Mossmann ist keine Neuauflage des früheren Werkes zum gleichen Thema, sondern ein völlig neues Buch. Der Inhalt ist wesentlich erweitert, da nunmehr alle Vertebraten einbezogen wurden und der umfangreiche Zuwachs an neuen Befunden, die in den vergangenen 50 Jahren erarbeitet wurden, nicht zuletzt durch die kontinuierliche Forschungsaktivität des Autors selbst, eingebaut wurde. Zahlreiche Lücken des Kenntnisstandes von 1937 konnten geschlossen werden, und Änderungen mancher theoretischer Deutungen wurden erforderlich. Das Werk ist ın erster Linie unter morphogenetischen und morphologischen Aspekten geschrieben und bietet eine zuverlässige und vollständige Informationsquelle zu Fragen der Frühentwicklung, einschließlich der extraembryonalen Anhangsorgane. Besonderer Nachdruck liegt auf der Lehre von der Placentation der Eutheria. Soweit irgend möglich, sind die physiologischen Grundprozesse berücksichtigt. Die sehr umfassende vergleichende Bearbeitung bildet eine gute Ausgangsbasis zur Erörterung evolutiver und phylogenetischer Fragen. Da das intrauterine Milieu bei Säugetieren wesentlich geringere artliche Differenzen erwarten läfst, als der Komplex äußerer Faktoren, denen der Gesamtor- ganısmus ausgesetzt ist, erkennt der Autor den Vorgängen der frühen Ontogenese einen hohen Erkenntniswert für stammesgeschichtliche Überlegungen zu. Grundsätzliche Überlegungen führen den Autor zur Neugliederung der Ontogeneseabläufe bei Wirbeltieren. Aufgrund der Beschaffenheit des Dottersackes und der Eistruktur unterscheidet MossMann: 1. Sessilata: Dottersack erscheint als Ausweitung des Darmkanals, ungestielt und an der Bildung der Bauchwand beteiligt. Eier: meso-megalecithal (Cyclostomata, Chondrostei, Holostei, Dipnoi, Teleostei, Amphibia). 2. Pedunculata: Dottersack gestielt und kaum an Bildung der definitiven Bauchwand beteiligt (Chondrichthyes). 3. Amniota: Amnion, Chorion und Allantois, Dottersack definitiv ganz oder zum großen Teil trılaminär. a. Sclerostraca: Eier megalecithal mit Schale, Schalenhaut und „Albumen“-Schicht: Reptilia, Aves, Monotremata. b. Hymenostraca: Eier miolecithal. Schalenhaut, Albumen und Zona pellucida vorhanden: Marsu- pialıa. c. Anostraca: Eier miolecithal, keine Schale und keine Schalenhaut. Selten eine dünne Albumen- schicht vorhanden: Eutheria. Das Werk enthält 77 Textabbildungen (Photos, EM-Photos, Diagramme) und 29 Tafeln, die jeweils ın einer Anzahl von sorgfältig ausgeführten, schematischen Strichzeichnungen wesentliche Entwick- lungsabläufe der verschiedenen Gruppen vergleichend darstellen. Das Buch kann nicht nur als zuverlässiges Handbuch und Nachschlagewerk dem Spezialisten wertvolle Dienste leisten, sondern ist wegen sehr klarer Sprache und didaktisch sorgsam überlegter Darstellungsweise auch hervorragend geeignet, jedem Biologen und Mediziner als Einführung zu dienen. Das Werk wird für lange Zeit seine grundlegende Bedeutung behalten und darf in keiner biologischen Bibliothek fehlen. Dem Autor gebührt Dank für diese Meisterleistung, dem Verlag für die ausgezeichnete technische Herstellung. D. STArck, Frankfurt/M. BEGon, M.; MORTIMER, M.: Population Ecology. 2. Ed. Oxford: Blackwell Scientific Publications 1986. 180 pp., 200 ıll. £ 10.50. ISBN 0-632-01443-1 Dieses Buch kann jenen empfohlen werden, die eine straffe und grundlegende Einführung in die Populationsökologie lesen möchten. Es wurden zoologische wie botanische Beispiele gleichermaßen berücksichtigt. Im Aufbau hat sich in der 2. Aufl. nichts wesentlich geändert. So werden auch jetzt die beiden Kapitel „Interspezifische Konkurrenz“ und „Prädation“ am umfassendsten behandelt. Sıe: spiegeln die englische Schule der Verhaltensökologie wider. Das Buch kann für Studierende, die sich in die Populationsökologie einarbeiten wollen, sehr hilfreich sein. R. SCHRÖPFER, Osmikrick NATUR + RECHT Zeitschrift für das gesamte Recht zum Schutze der natürlichen Lebensgrundlagen und der Umwelt NATUR + RECHT ist im deutschen Sprachraum die erste spezielle Zeitschrift für das gesamte Recht zum Schutze der natürlichen Lebensgrundlagen. Sie willdem Naturschutz- und Landschaftspflegerecht in Wissenschaft, Ausbildung und Praxis zu der Bedeutung verhelfen, die ihm als Kernstück des Umweltschutzrechts zukommt. NATUR + RECHT macht deshalb die Gemeinsamkeiten landesrechtlicher Sonderregelungen transparent und faßt die Publikationen auf verwandten Rechtsgebieten zusammen, die sich auf eine Vielzahl von Rechts- und interdisziplinären Zeitschriften mit häufig nur regionaler Bedeutung verteilen. NATUR + RECHT setzt als Rechtszeitschrift die sachliche Diskussion in Gang und belebt sie, verbessert Kommunikation und Argumentation und trägt somit zur Rechtsfortbildung bei wie auch zum Aufbau des oft beklagten Vollzugsdefizits. « NATUR + RECHT. Zeitschrift für das gesamte Recht zum Schutze der natürlichen Lebensgrundlagen und der Umwelt. Herausgegeben von Claus Carlsen, Dieter Engelhardt und Ludger Pielow. Erscheinungsweise: 8 Hefte pro Jahr. Abonnementspreis (1988): jährlich 248,— DM zzgl. 10,50 DM Versandkosten BPAJL (Inland), bzw. 12,— DM (Ausland). Einzelheft 34,— DM Verlag Paul Parey : Hamburg und Berlin P/REV Erscheinungsweise und Bezugspreis 1988: 6 Hefte bilden einen Band. Jahresabonnement Inland: 308,- DM zuzüglich 13,80 DM Versandkosten, Gesamtpreis 321,80 DM einschließlich 7% Mehrwert- steuer. Jahresabonnement Ausland: 308,— DM zuzüglich 18,- DM Versandkosten. Das Abonne- ment wird zum Jahresanfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, Spitalerstraße 12, D-2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesellschaft für Säugetier- kunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. Z. Säugetierkunde 53 (1988) 5, 257-320 Eın Paradies darf nıcht sterben! Fesselnder Reisebericht aus dem Regenwald D: grüne Hölle - legendenumwobenes bedrohtes Tier- und Pflanzenparadies in Mittel- und Südamerika ist 20mal so groß wie die Bundesrepublik Deutschland. Durch die Zerstörung der tropischen Regenwälder ist langfristig auch das Klima in Mitteleuropa bedroht. Mehrere Studienreisen führten Jänos Regös, einen Schweizer Biologen und Ökologen, nach Costa Rica, Brasilien und Peru, wo er die Tier- und Pflanzenwelt des Urwaldes erforschte: Er schildert seine ungewöhnlichen Begegnungen mit Schlangen, Affen, Fröschen, Eidechsen, Vögeln und anderen Tieren und beschreibt die Flora der Regenwälder mit ihren auch heute noch teilweise unerforschten Bäumen, Flechten, Moosen, Farnen, Bromelien und Orchideen. Von seinen Reisen brachte der Autor eindrucksvolle Fotos mit. Sie zeigen die ganze Schönheit aber auch die fortschreitende Zerstörung dieses einzigartigen Paradieses. Zu beziehen durch jede Buchhandlung. % Jänos Regös. Die grüne Hölle - Ein bedrohtes Paradies. Bericht aus dem Regenwald. 1987. 130 Seiten und 64 Tafeln mit 147 farbigen Abbildungen und 11 Zeichnungen ım Text. Gebunden 39,80 DM eye Verlag Paul Parey - Hamburg und Berlin 'ol. 53 (6), 321-384, Dezember 1988 ISSN 0044-3468 C21274F 50 | =» ITSCHRIFT FÜR SRUGETIERKUNDE NTERNATIONAL JOURNAL JF MAMMALIAN BIOLOGY rgan der Deutschen Gesellschaft für Säugetierkunde ıportant Publisher’s Notice — Wichtiger Hinweis des Verlages | lettaz, R.; Aulagnier, S.: Statut de trois especes de chiropteres rares au Maroc: Nycteris thebaica, Hipposideros caffer, et Pipistrellus rueppelli. — Status of three rare bat species in Morocco: Nycteris thebaica, Hipposideros caffer and Pipistrellus rueppelli. -— Zum Vorkommen von drei seltenen Fledermausarten in Marokko: Nycteris thebaica, Hipposideros caffer und Pipistrellus rueppelli 321 the, H.; König, A.; Radespiel, Ute; Darms K.; Sieß, Margaretha: Occurrence and frequency of twin-fight in the Common marmoset (Callithrix jacchus). — Auftreten und Häufigkeit von Zwillingskämpfen beim Weißbüscheläff- chen (Callithrix jacchus) 325 vet, J.; Dolivo, M.; George, C.; Gogniat, A.: Homing behavior of Wood mice (Apodemus) in a geomagnetic anomaly. — Heimfindeverhalten von Wald- und Gelbhalsmäusen (Apodemus) in einer geomagnetischen Ano- malie 333 üller, E. F.; Soppa, U.: Activity pattern and thermoregulation in the Cuis (Galea musteloides Meyen, 1833). — Aktivitätsmuster und Temperaturregulation beim Wieselmeerschweinchen (Galea musteloides Meyen, 1833) 341 ck, W. F.: Die Bedeutung des Untergrundes für die Größe von Bauen des Dachses (Meles meles) am Beispiel zweier Gebiete Südostbayerns. — The importance of soil composition for the size of badger setts (Meles meles), in two areas in SE Germany 349 ıker, Carolyn M.: Scent marking behaviour in captive Water mongooses (Atilax paludinosus). — Duftmarkierung bei gefangengehaltenen Sumpfichneumons (Atilax paludinosus) 358 scher, M. S.: Zur Anatomie des Gehörorganes der Seekuh (Trichechus manatus L.), (Mammalia: Sirenia). — Contributions to the anatomy of the hearing organ of the seacow (Trichechus manatus L.) (Mammalia: Sirenia) 365 issenschaftliche Kurzmitteilung iller, T. K.; Biknevicius, A. R.; Kat, P. W.: Home range of an African wildcat, Felis Silyestris (Schreber), near Elmenteita, Kenya. _ Streifgebiet einer afrikanischen Wildkatze Felis silvestris (Schreber) nat ne Elmenteita, Kenia 380 382 384 akanntmachungen ichbesprechungen /erlag Paul Parey Hamburg und Berlin HERAUSGEBER/EDTITORS P. ]. H. van BREE, Amsterdam - W. FIEDLER, Wien - H. Frick, München - W. HERRE, Kiel - H.-G. Krös, Berlin - H.-J. Kunn, Göttingen - E. KuLzer, Tübingen - B. Lanza, Florenz — J. NIETHAMMER, Bonn - H. ReıcHstein, Kiel - M. RöhHrs, Hannover — D. STARCK, Frankfurt a. M. - F. Strauss, Bern — E. THenIıus, Wien - P. VoceL, Lausanne SCHRIEFETLELTTUNG/EDITORSIAT OFLIEE H. SCHLIEMANN, Hamburg —- D. Kruska, Kiel This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con- tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information Die Zeitschrift für Säugetierkunde veröffentlicht Originalarbeiten und wissenschaftliche Kurzmittei- lungen aus dem Gesamtgebiet der Säugetierkunde, Besprechungen der wichtigsten internationalen Literatur sowie dıe Bekanntmachungen der Deutschen Gesellschaft für Säugetierkunde. Verantwort- licher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Harald Schliemann. Zusätzlich erscheint einmal ım Jahr ein Heft mit den Abstracts der Vorträge, die auf der jeweiligen Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde gehalten werden. Sie werden als Supplement dem betreffenden Jahrgang der Zeitschrift zugeordnet. Verantwortlich für ihren Inhalt sind ausschließlich die Autoren der Abstracts. Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z. Hd. Prof. Dr. Dieter Kruska, Institut für Haustierkunde, Biologie-Zentrum, Neue Universität, Olshausenstr. 40-60, D-2300 Kiel. Für die Publikation vorgesehene Manuskripte sollen gemäß den „Redaktionellen Richtlinien‘ abgefaßt werden. In ihnen finden sich weitere Hinweise zur Annahme von Manu- skripten, Bedingungen für die Veröffentlichung und die Drucklegung, ferner Richtlinien für die Abfassung eines Abstracts und eine Korrekturzeichentabelle. Die Richtlinien sind auf Anfrage bei der Schriftleitung und dem Verlag erhältlich. Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der Korrekturfahnen erfolgen. Vorbehalt aller Rechte: Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei- ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch ım Einzelfall grund- sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch ım Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber- rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of the first page of an article in this journal indicates the copyright owner’s consent that copies of the article may be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective, or for resale. For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the Mit zwei Beilagen des Verlages Paul Parey Fortsetzung 3. Umschlagseite © 1988 Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstr. 12, D-2000 Hamburg 1; Lindenstr. 44-47, D-1000 Berlin 61. — Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., D-2240 Heide/Holst. Important Publisher’s Notice Duetothe development ot internationaleconomy sincethe early eighties, the acquisition budgets of scientific libraries all over the world often were drastically reduced. At present there is no end of this tendency in sight — a change for the better even less. The libraries are compelled to answer with a substantial reduction of their journal subscriptions. As a result the edition-runs of individual journals decrease, too. By decreasing edition-runs, the cost per copy increases and consequently the retail price — more cancellations and finally economic failure will follow. The individual user has not yet recognized the situation because, in many cases, he does not use the original journal but gets a photocopy of the original from the library. We have to realize, however, that the scientific research journal as the central medium for documentation of research results and international information exchange cannot be replaced today or ın the near future by any other medium. Neither micro-film archives nor electronic non-print media are alternatives. If you have followed our line of reasoning thus far you must agree that in order for scientific research journals to survive, basic economic requirements must be organized. The easiest way would be, as explained, for libraries to be prevented from cancelling their subscriptions. Our appeal today to you personally is to approach your librarian as soon as possible and explain that the availability of the journal at your library ı is essential for the efficacy of your research work and the broad education of the rising generation of scientists. We do hope that our joint efforts will be successful. Paul Parey Scientific Publishers Wichtiger Hinweis des Verlages Aufgrund der Entwicklung der Weltwirtschaft seit Anfang der 80er Jahre wurden die Anschaffungsetats der wissenschaftlichen Bibliotheken international teilweise mehrfach drastisch verringert. Ein Ende dieses Trends ist gegenwärtig nicht erkennbar - erst recht nicht eine Umkehr. Die Bibliotheken müssen, der Not gehorchend, mit einer wesent- lichen Reduzierung der Zahl der abonnierten Zeitschriften reagieren. Damit sinkt auch die Verkaufsauflage der einzelnen Zeitschriften, die von Abbestellungen betroffen sind. Bei sinkender Verkaufsauflage steigen die Einzelstückkosten der Zeitschrift und damit die Preise - weitere Abbestellungen und damit der wirtschaftliche Zusammenbruch der Zeitschrift sind die Folge. Der einzelne Benutzer hat diese Situation bisher kaum erkannt, da er ın vielen Fällen schon heute nicht die Original-Zeitschrift benutzt, sondern von seiner Bibliothek anstelle des Original-Exemplars Xerokopien erhält. Man muß sich aber bewußt machen, daß die wissenschaftliche Forschungszeitschrift als das zentrale Medium für Befund-Dokumentation und internationalen Informations- austausch weder heute noch in absehbarer Zukunft durch andere Medien ersetzbar ist. Weder Mikrofilm- Archive noch elektronische Non-Print-Medien sind eine Alternative. Wenn Sie unserer Argumentation soweit gefolgt sind, werden Sie uns zustimmen, daß auch die ökonomischen Voraussetzungen für ein Überleben der wissenschaftlichen Forschungszeitschriften geschaffen werden müssen. Dies läßt sich am einfachsten erreichen, wenn — wie erläutert - verhindert wird, daß Zeitschriften durch Bibliotheken abbestellt werden. Unser heutiger Aufruf an Sie ganz persönlich geht dahin, daß Sie möglichst bald mit Ihrem Bibliothekar sprechen und deutlich machen, daß die schnelle Verfügbarkeit der Zeitschrift innerhalb Ihrer Bibliothek für die Effektivität Ihrer Forschungsarbeit und die vielseitige Ausbildung des wissenschaftlichen Nachwuchses von ausschlaggebender Bedeutung ist. Wir hoffen auf einen Erfolg unserer gemeinsamen Bemühungen. Verlag Paul Parey Z. Säugetierkunde 53 (1988) 321-324 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Statut de trois especes de chiropteres rares au Maroc: Nycteris thebaica, Hipposideros cafter et Pipistrellus rueppelli Par R. ArLETTAZ et S. AULAGNIER Musee d’histoire naturelle, Fribourg, Suisse et Departement de biologie, Universite Cadi Ayyad, Agadır, Maroc Reception du Ms. 15. 10. 1987 Abstract Status of three rare bat species in Morocco: Nycteris thebaica, Hipposideros caffer and Pıpistrellus rueppelli Discussed is the occurrence of three rare species of bats in Morocco. Nycteris thebaica and Flipposideros caffer originate from Africa, south of the Sahara; in Morocco they live exclusively along the atlantie coast. Pipistrellus rueppelli which was caught in che presaharian area (Tafılalt) is reported for the first time in Morocco. The populations of all three species are considered as remnants. Introduction Au Maroc, les &tudes mammologiques ont connu deux p£riodes d’intense activite: !’une au debut du siecle dont les resultats sont rassembles dans l’ouvrage de CABRERA (1932), ’autre dans les annees cinquante avec les travaux de Panouse (1951, 1958, 1959) et de BRossET (1960). Mais les prospections d’alors se sont surtout limitees au Maroc septentrional alors que les regions situees au sud du Haut-Atlas restaient meconnues. Cette lacune tend A &tre comblee progressivement par les naturalistes residents et par les visiteurs occasıonnels, attires par les paysages desertiques du Maroc pre&saharien. L’observation de trois especes de chiropteres peu communs s’inscrit dans cette dynamique. Materiel et methode Nos observations r&sultent essentiellement de la visite diurne des grottes karstiques et autres cavites souterraines; quelques captures directes ont EtE effectuees A cette occasion. Accessoirement, nous avons place des filets de nylon sur des plans d’eau calme (flaques residuelles d’oueds, bras de rivieres, etc.) ainsı qu’a l’entree d’un conduit de canalisation abandonnee. Resultats Nycteris thebaica Geoffroy, 1818 - Nyctere de la Thebaide Decouverte assez recemment dans la proche region de Rabat (Panouse 1958), cette espece d’origine Ethiopienne semble repandue sur toute la facade atlantique marocaine. Trouvee peu de temps plus tard dans l’Anti-Atlas occidental (PanouseE 1959), elle a aussi ete collectee sur le versant septentrional du Haut-Atlas par la Smithsonian Institution, en 1971. La region d’Agadir heberge egalement, au moins temporairement, de petites colonies de Nycteris: ’une, forte d’une cinquantaine d’individus, fut decouverte pres d’Imouzzer des Ida Ou Tanane, en octobre 1985, dans une petite grotte (observation non renouvelee depuis en depit d’un suivi regulier des cavites de la zone); l’autre, reduite a une dizaine U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5306-0321 $ 02.50/0 322 R. Arlettaz et S. Aulagnier d’individus, fut observee en avrıl 1986 et janvier 1987 dans une canalisation abandonn&e pres de l’embouchure de l’oued Massa. Toutes les donnees relatives A cette espece proviennent de cavites aux dimensions reduites; dans les deux derniers cas, la hauteur de la galerie n’excede pas un metre. La temperature peut Etre tres variable; oscillant entre 15° et 20°C durant la periode ALGERIE Localıtes citees dans le texte: 1 = Tanger; 2 = Rabat; 3 = Essaouira et Mogador; 4 = Imouzzer des Ida Ou Tanane; 5 = Agadır; 6 = Oukaimeden; 7 = Foum Zguid; 8 = Aoufouss (Tafilalt); 9= Abadla; 10 = Beni Abbes; a = oued Cherrat; b = oued Souss; c = oued Massa; d = oued Ziz. © = stations de Pipistrellus rueppelli (quatre mentions pour l’Afrique du nord-ouest) d’occupation (hiver) de la grotte de l’oued Cherrat (Panouse 1958), elle etait de plus de 22 °C (automne) et largement superieure A 26 °C (printemps) pour les dernieres observa- tions en date. Actives toute l’annee, ces chauves-souris semblent se deplacer et changer de gite frequemment. Leur Ecologie demeure tres floue (aucune colonie de parturition n’a Ete decouverte A ce Jjour...). Hipposideros caffer (Sundevall, 1846) - Rhinolophe de Cafrerie Ce «Rhinolophe» est connu au Maroc depuis le debut du siecle losque CABRERA (1906) decrivit l!’espece Hipposideros tephrus, sur la base d’un specimen provenant de la region d’Essaouira. Les vues des taxonomistes divergent A cet Egard, la plupart attrıbuent cependant ä la population maghrebine le rang subsp&cifique adopte ıcı (Hipposideros caffer thephrus). CABRERA (1932) considere cette forme comme repandue «desde Täanger hacıa el Sur (localidad tipica, Mogador), hasta el Senegal»; pourtant, aucune observation n’est venue actualiser ces donnees jusqu’a ces derniers temps. La premiere decouverte (etonnante!) fut celle d’un cadavre A l’Oukaimeden (Haut- Atlas, 2500 m). Mais la confirmatıon de la presence de ce chiroptere au Maroc fut apportee par la capture, en ayrıl 1985, de quatre mäles ä l’entree d’une canalısation abandonnee A P’embouchure de l’oued Massa. Ce gite Etait rEoccupe au printemps suivant par une colonie d’une vingtaine d’individus (un mäle capture, avant-bras = 45 mm) et dans le courant de l’hiver par une quinzaine d’animaux tres actifs. Aınsi, A plus de cinquante ans d’intervalle, ıl est etabli que la facade atlantique marocaine heberge cette espece tres repandue en Afrıque au sud du Sahara. Statut de trois especes de chiropteres rares au Maroc 325 Pipistrellus rueppelli (Fischer, 1829) — Pipistrelle de Rüppel Le 19 avrıl 1985, parmi troıs Pipistrellus kuhli (deux femelles et un mäle), une femelle de cette espece Etait capturee au filet au-dessus d’un bras mort de l’oued Ziz, A Aoufouss (Tatilalt). Cette magnifique Pipistrelle se distingue d’embl&ee des autres representants de son genre par la coloration de son pelage (poils du dos gris argentes, face inferieure blanc creme) et par la forme de ses incisives internes sup£rieures qui sont profond&ment bifides (Haymann et Hırr 1971). La peau du visage et des oreilles est d’un brun tres fonce, presque noire; les membranes alaires sont egalement gris-brun sombre. Les principales mensurations de cet anımal sont les suivantes: avant-bras = 32.4 mm; 4£me doigt = 50.3 mm; 5eme doigt = 43.6 mm. Ce chiroptere est le seul representant du sous-genre Scotozous, Dobson 1875, vivant sur le continent afrıcain. La decouverte de cette chauve- souris apporte un complement a l’etude des chiropteres de cette region qui est une zone de transition entre domaines palearctique et saharıen (AULAGNIER et DESTRE 1985), mais aussi a la connaissance de la faune marocaine avec une vingt-sixieme espece de chauve-souris. Apres verification, ıl s’avere que la Smithsonian Institution a Egalement collecte un specımen de P. rueppelli dans les environs de Foum Zguid, le 10 mars 1970 (femelle - 0470693), au cours de sa gigantesque campagne de prospection 1969-1972; cette observa- tion est unıquement mentionnee dans le registre des collections de cet organisme. La capture filali constitue donc la seconde mention au Maroc pour cette espece d’origine africaine qui parait rare au nord du Sahara. Discussion Outre leur relative rarete, un trait commun ä& ces trois especes est leur large r&partition afrıcaine qui les oppose aux autres chauves-souris qui peuplent le Maroc et qui sont palearctiques pour la plupart, sahariennes pour les tormes localisees dans les zones les plus meridionales. De fait, ıl s’avere que Nycteris thebaica er Hipposideros caffer frequentent essentiellement la facade atlantique marocaine, en particulier la plaine du Souss, comme certains autres mammiferes qui constituent des relictes de faune tropicale: Crocidura viaria, Xerus erythropus, Mastomys erythroleucus Mellivora capensıs, etc. La distribution de Pipistrellus rueppelli est sensiblement differente et plus complexe ä interpreter: largement repandue en Afrique australe mais rare en Afrique de l’ouest, les donn&es marocaines s’ajoutent ä deux mentions algerıennes (Beni Abbes [Haymann et Hırr 1971] et Abadla [Gaisler et Kowarskı 1986] - localıtes distantes repectivement de 250 et 150 km d’Aoufouss), etablissant l’existence d’un peuplement sur la frange septentrionale du Sahara. Il reste A poursuivre les investigations pour pre£ciser le statut de ces chauves- souris et ainsi apprehender les caracteristiques de leurs populations qui semblent relic- tuelles. Remerciements Nous remercions G. DÄNDLIKER, H. DUPERREX, J. L. ROLANDEZ, J. P. MARFIN, L. LEsneE et M. THEVENOT qui ont particip€ aux recherches sur le terrain, le Professeur V. AELLEN, Directeur du Museum de Gen£ve, qui a relu et critiqu& le manuscrit, enfin A. SCHUBERT, K. ZBINDEN et P. ZıingG pour la traduction des resumes. Resume- Nycteris thebaica et Hipposideros caffer sont des chiropteres d’origine africaine, frequents au sud du Sahara; au Maroc, ils n’ont et trouves que le long de la cöte atlantique. Pıpistrellus rueppelli, capture dans la zone presaharienne (Tafilalt), est mentionne pour la premiere fois au Maroc. Les populations de ces trois especes peuvent Etre considerees comme relictuelles. 324 R. Arlettaz et S. Aulagnier Zusammenfassung Zum Vorkommen von drei seltenen Fledermausarten in Marokko: Nycteris thebaica, Hipposideros caffer und Pipistrellus rueppelli Das Vorkommen von drei seltenen Fledermausarten in Marokko wird diskutiert. Nycteris thebaica und Hipposideros caffer stammen aus Afrika, südlich der Sahara. In Marokko leben sie ausschließlich entlang der atlantischen Küste. Pipistrellus rueppelli konnte ım Südosten des Landes (Tafilalt) gefangen und damit zum erstenmal in Marokko nachgewiesen werden. Bei allen drei Fledermausarten dürfte es sich um Restpopulationen handeln. Bibliographie AULAGNIER, $.; DESTRE, R. (1985): Introduction a l’etude des chiropteres du Tafılalt (sud-est marocain). Mammalia 49, 329-337. BrossET, A. (1960): Les mammiferes du Maroc oriental. Leur r&partition, leur statut actuel. Bull. Soc. Sc. nat. phys. Maroc 40, 243-263. CABRERA, A. (1906): Mamiferos de Mogador, Bol. Real. Soc. esp. Hist. nat. 6, 357-368. — (1932): Los mamiferos de Marruecos. Trab. Mus. nac. Cienc. nat., ser. zool. 57, 1-361. GAISLER, J.; KOwALsKT, K. (1986): Results of the netting of bats ın Algerıa (Mammalıa: Chiroptera). Vest. cs. Spolec. zool. 50, 161-173. Haymann, R. W.; Hırı, J. E. (1971): Order Chiroptera. Part 2, 73 p. In: The Mammals of Africa: an identification manual. Ed. J. MEEsSTER and H. W. SETZER. Washington: Smithsonian Institution Press, PAnouse, J. B. (1951): Les chauves-souris du Maroc. Trav. Inst. Sc. Cherif. 1, 1-120. — (1958): Presence au Maroc de Nycteris thebaica (Chiroptera). Bull. Soc. Sc. nat. phys. Maroc. — (1959): Une nouvelle localiıt€e marocaine pour Nycteris thebaica. C. R. Soc. Sc. nat. phys. Maroc 255.82. Adresse des auteurs: RAPHAEL ARLETTAZ, CH-3960 Loc, Suisse; STEPHANE AULAGNIER, Departe- ment de biologie, Universite Cadı Ayyad, B. P. 28/S, Agadir, Maroc Z. Säugetierkunde 53 (1988) 325-332 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Occurrence and frequency of twin-fight in the Common marmoset (Callithrix jacchus) By H. RoTHE, A. König, UTE RADESPIEL, K. DARMS und MARGARETHA SIESS Institute of Anthropology, University of Göttingen Receipt of Ms. 9. 6. 1987 Abstract Studied frequency and occurrence of twin-fight in the common marmoset (Calhthrix jacchus). Twin- fight is found in 32 of 39 groups and in 27 of 83 litters. 39 twins are heterosexual, 23 isosexual female and 16 isosexual male. Twin-fights are most frequent in isosexual male (50 %) and female (46 %) litters, least frequent in male-female twins (17 %). The age of the combatants at twin-fight is on the average 188,5 days in mm-twins, 241,7 days in ff-twins and 237,1 days in mf-twins. Sibling competition ıs observed during the whole interbirth-interval of the a-female. Five twin-fights show a temporal relation to the first and second estrus post partum of the a-female. In 6 % of all twin-fights a combatant has to be removed in order to prevent severe injuries. Sibling competition is observed in groups of max. 10 members. A relatively high percentage of twin-fights (ca. 37 %) is noticed in very small groups and in absence of group members of the opposite sex (parents excluded). Introduction Agonistic interactions between infantile/juvenile social living mammals are the exception rather than the rule. The siblings of a given litter, for example Suidae, Canıdae, Felidae or Cricetidae often quarrell for the access to the nıpples of their mother or for food, and also during play sessions they may interact to some extent agonistically. However, all these interactions do not cause injuries to their famılıar social partner (see also SUTCLIFFE and PooLE 1984). Poor management conditions (e.g. overcrowding, stimulus deprived environment) often induces increased aggressiveness between littermates which may lead to severe physical consequences, and even to the death of one or the other sıbling (e.g. cronism in piglets or hamsters). All these dissociative interactions are characterized insofar as they are trıggered by an actual event, for example the access to food or mother’s nipples. Furthermore they do not seem to have negative longterm effects on the relationship of the combatants. On the contrary many authors have stressed the eminently associative character of the interactions between infantile mammals (for a survey see FAGEN 1981). In the behavioural ontogeny of the marmosets however, we can observe a process which does not seem to be promoted by an actual event - at least we could not detect ıt - which exclusively involves dissociative behaviours, and which often ends in severe physical, and possibly even psychic injuries in the 4-10 months old marmoset twins. This twin-fight (SurcLirre 1980) or sibling competition (KrEıman 1979) is believed to deter- mine the relative position of the twins in the hierarchy of the group. According to SUTCLIFFE and PooLE (1984) the twin-fight has longterm consequences on the hierarchical relationship of the siblings, that is, the dominance-subordination- relationship between the twins will not be altered as long as the twins live together in their U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5306-0325 $ 02.50/0 H. Rothe, A. König, Ute Radespiel, K. Darms und Margaretha Sieß 326 -ure2199un A], = ( ) ‘posresı-puey = YH ‘Joyrouı 199507 Aq paJeaı sI9ad = WH (TE) €E €8 (T)+9S jeı0ı z (br) 09 (1)+6 wwxz (NH) wu + uw (nd) u + u (8E) Hr (1)+6 wu 9 el HZ (NH) yuu br al H 21 PE zug I (yH) zu j (N) gu 8I ze zu uUONEI[2IsU09 um] Dell I 4J,L 3NnoyyAM pur YUMA SIOIJTT Jo samuanba.F IInNJosge Pur JAHEIIY "T olgPL "DOPNIIXa due S199d P919I507 pur poresıpuey ‘umoyun Jussed Juo Jo uONeIIUId = x !ure1199un IyS1J-UIMI = BOpz] p JS pLESD PuEH a1 } ey (1)-+7 (D)+#S (1)- AL (1)+ A,L OU (D-/ (1)- En? (1)+6 (D+I AL OU Ol SlblE el ‚LL ou X all ce 9 ‚LL ou yuu & ae ee ER ad ra/d r1/2d 1/4 ad uone[jasuoI UIMI IyS7-uIm? InoyyIMm sıayjı] - LI 0U HySyy-umy yaMm (S)13931] — q,L :Aduanbary-L pure sum jo uoneIsusn) ‘7 9]qYL, Twin-fight in the Common marmoset 927, natal group. KLEIMAN (1977) does not believe that marmoset families are hierarchically structured (compare however ErrLE 1975; ROTHE 1975, 1979; STEvENSON and POOLE 1976). On the other hand KrEıman (1979) refers to dominant and subordinate twins ın Leontopithecus rosalıa rosalıia. In the present paper we give data on the occurrence, frequency and the relationship of twin-fights to group size, sex ratio of the natal group, age at twin-fight, as well as the incidence of twin-fights during the interbirth-interval of the &-female. Material and methods 83 litters (5 hand- and fostermother-reared peers included, see Table 2) of 32 families of our Callıthrix jacchus colony could be analysed. In 39 litters the surviving siblings were bisexual, in 23 isosexual female and in 16 isosexual male. The size of the groups (parents and offspring) varied from 4 to 18 members. The generation of the twins/peers was F1 to F3F5/F3 (Table 1). The data were taken from the diary of our primate laboratory, in which we record all important biological and behavioural events which can be observed during the anımals’ daily activity (6-18 h). The groups were housed in cages or rooms of 1.0mx2.0mx2.5m to 5.0mx 7.0 mx3.0m in size, each of them being equipped with free-swinging climbing frames, feeding boards and sleeping boxes. The anımals usually could not see each other, however occasionally there was some acoustic and olfactory contact. In addition to daylight artificial lighting was provided by neon tubes on a 12 hour cycle (6-18 h), and the rooms were screened by venetian blinds from 5.00 p.m. to 7.00 a.m. A constant temperature of 26 °C and a humidity of 70 % was maintained by means of an air conditioner. The animals were fed twice daily. Most of the twin-fights (TF) were not observed directly, that is, we do not know the initiator and the special circumstances which triggered the sıbling competition. From the numerous small wounds, which could be detected in the face and in other parts of the twins’ bodies as well as from their aggressive interactions it was rather easy to infer, that TF had taken place. We cannot exclude, however, that we have overlooked one or the other TF, especially those which did not cause injuries or which were not accompaniıed by detectable aggressive interactions between the combatants. Results TF were noticed in 22 of the 32 groups (68.68 %) and ın 27 of the 83 litters (32.5 %) (Table 2). In 3 groups we observed 2, in one group 3 TF (see also KrEıman 1979 for Leontopithecus rosaha rosalia). Most frequent TF occurred ın isosexual male litters (= mm-litters) (n = 9, 50.0 %). Isosexual female twins (= ff-litters) (n = 24) had also a high rate of TF (n = 11, 46 %), whereas male-female twins (mf-litters) (n = 41) had significantly lower TF-ratios (n = 7, 17 %). According to these data TF was observed in about one third of all litters. From these data we can conclude that TF ıs not a regular event in the ontogeny of the common marmoset infants/jJuveniles. In 5 TF the loser was so severely attacked and injured by its sıbling that he has to be removed from its family, one of them died 4 days following the TF (see also KLeıman 1979 for Leontopithecus rosalıa rosalia). In our C. jacchus colony TF was observed when the combatants were 4 to 10 months old. The maximum TF-age is rather ıdentical in the three twin-constellations (see also SUTCLIFFE and PooLE 1984). The age of the combatants at TF was lowest in mm-twins (on the average 188.5 days, range 112-278 days). In ff-twins we find the highest TF-age (on the average 241.7 days, range 167-314 days), closely followed by mf-litters (on the average 237.1 days, range 209-314 days) (Table 3). The low TF-age of the mm-twins is mainly due to 2 TF which were observed when the combatants were only 4 months old. As ıs demonstrated ın the Figure TF are distributed over the whole interval between 2 births by the &-female of the group. The peak however lies in the first half (= 50-80 days) 328 H. Rothe, A. König, Ute Radespiel, K. Darms und Margaretha Sieß Table 3. Age of twins at twin-fight (in days) twin constellation range mf 237 209-314 ff 41. 167-314 mm h 112-278 Table 4. Relationship of mm-TF-frequency to sex ratio of the family; parents and following litter are excluded 2 DON ON WI RWN Hm 2X T = twin-fight; E = twin-fight related expulsion/removal; X = no twin-fight; ( ) = twin-fight uncertain; a= reared by foster mother; b = handreared Table 5. Relationship of ff-twin-fight-frequency to sex ratio of the family (See legend of Table 4) oOoNON UV RUDMTO Table 6. Relationship of mf-twin-fight frequency to sex ratio of the family See also legend of Table 4 0 1 2 3 4 5 6 7 8 9 Twin-fight in the Common marmoset 329 31-40 11-20 0-10 21-30 41-50 61-70 81-90 101-110 121-130 41-50 21-30 0-10 11-20 31-40 51-60 71-80 91-100 111-120 days preceding days following birth birth Incidence of TF in relation to preceding/following birth in the family group. n = frequency of TF; —_ WE uncertain of the «-temale’s pregnancy. The TF in the three twin-constellations are rather similar distributed over the interbirth-interval. Five TF however showed a temporal relation to the first and second estrus post partum of the a-female (days 11-37), the twins being 167 to 314 days old at that time. TF which required the removal of a combatant occurred only in isosexual litters [n = 5; 3 twins, 2 handreared peers (male and female)]. If we exclude the handreared peers, then 12.5 % ot all expulsions/removals of a group member followed a TF, and in only 6 % of all TF (n = 83) we had to remove a sibling in order to prevent severe injuries or even the death of the inferior twin. Six from 9 TF of mm-twins and -peers were observed when the sex ratio of the group was in favour of males, 2 when the sex ratio was balanced, and only 1 when the females were in surplus. TF independent expulsions of group members could only be noticed when males were in surplus (see Table 4). We obtain the same results for ff-sıblings. Ten of 11 TF were seen when the sex ratio was in favour of females, only 1 when the group had a balanced sex ratio. Most of the TF were observed in groups with 6 to 10 members. It must be stressed however, that we did not see any TF ın groups with extremely unbalanced sex ratio. Wether this depends on the unbalanced sex ratio or on the group size or on both, cannot be answered (see Table 5). TF in mf-siblings could nearly exclusively be observed in groups with rather balanced sex ratio and with less than 10 group members (see Table 6). In summing up we get the following results. In our C. jacchus colony we observed TF in groups of max. 10 group members. A relatively high percentage (approx. 37 %, n = 10, 8 in ff-twıns) of TF was noticed in very small groups and in the absence of group members Table 7. Relationship of twin-fight frequency to sex ratio of the family Summary of Table 4-6 ) 1 2 3 4 5 6 7 8 9 330 H. Rothe, A. König, Ute Radespiel, K. Darms und Margaretha Sieß of the opposite sex (parents excluded). In that situation TF often led to the removal of a combatant. Between these two extreme situations it seems to give a stable group size with neither TF nor expulsions/removals (see Table 7). Discussion According to the data of our C. jacchus colony sibling competition is not a regular event in the relationship between young common marmosets. Even if we assume that many TF could not be observed, there still remained a large number of litters in which TF did not occur. The infants of wildcaught parents showed the least TF. With increasing number of generation the frequency of TF rises, but not continuously. It might be, however, that with increasing length of the existence of the colony in our laboratory a better monitoring ot the anımals was realized so that more TF could be detected compared to the first years of colony existence. SUTCLIFFE and PooLE (1984) argue that TF have a longterm effect of the relative hierarchy between the combatants. We cannot confirm this view, since TF are not seldomly repeated and the rank position may be changed (see also Köntc, in prep.; Kreıman 1979 for Leontopithecus rosalıa rosalia). TF-related expulsions are rather seldom. Since the age at TF ıs relatively low (see also SUTCLIFFE and PooLE 1984; KLEıman 1979 for Leontopithecus |10-12 months]; WOLTERS, pers. communication for Saguınus oedıpus oedipus |8-12 months]), this means, that the animals are far from being adult (see ABBoTT and HEARN 1978) ıt seems biologically meaningful to prevent expulsions in order not to reduce too drastically life expectancy and/or reproductive success of the infantile/juvenile loser. According to SUTCLIFFE and PooLE (1984) TF must take place at an age when the permanent dentition has not yet developed to avoıd dangerous injuries. In most TF which were noticed in our colony only minor wounds indeed occurred, but in some TF also severe injuries could be observed ın both combatants (see also Krerıman 1979 for Leontopithecus rosalıa rosalia). As yet we have no idea on the psychic effects/consequences ot TF which might also be, even more important, than any physical wound. Most remarkable is the sudden appearance - at least for the human observer - of TF. The initiating event was not observed in most cases. SUTCLIFFE and POOLE (1984) believe that the increase of intolerant behaviour of older group members towards the twins could stimulate the siblings to TF. KreEıman (1979) mentions as possible releaser for TF in Leontopithecus rosalıa rosalıa the fırst transfer of the youngest family members (= 2 weeks old) from the mother to the father. For both assumptions we do not have any indication from C. jacchus. Without further information on the initiating event it is somewhat difficult to understand that aggressiveness toward the twins should trigger more or less suddenly aggressiveness between them. We urgently need more data on this aspect. According to AssortT (1978) C. jacchus already show from the sıxth month of age considerable changes in the estradiol (female) and testosterone (male) levels. These data fıt relatively well to the TF-age. Therefore ıt might not be unrealistic to assume that the endocrine status of the twins could influence, and may be the primary cause for TF. If this is the case then TF would be a regular event in the development of common marmosets and we had overlooked a considerable number of TF in our colony. It may be, however, that many TF proceed less spectacularly, i.e. in a strong ritualized, non-fighting manner, so that they are hardly to detect. Our data show that TF are influenced by the sıze and the sex ratio of the famıly. It ıs most striking, that TF very often occur in small groups, that is in groups in which besides the parents no other or only very few adult members live. According to our observations these groups are only scarcely hierarchically structured. The probability to collide with an adult brother or sister is essentially lower in small than in large groups in which TF Twin-fight in the Common marmoset 3a typically occur seldom or not at all. This result does not fit to SUTCLIFFE and POOLES (1984) hypothesis on the ınitiating event/anımal of TF. At least some TF seem to be triggered by the estrus of the «-female, especially by the post-conception-estrus ın mid-pregnancy. KrEıman (1979) also observed in Leon- topithecus rosalia rosalia estrus related TF and parent-offspring-conflicts during mid- pregnancy (mid-pregnancy false estrus sensu KLEIMAn and Mack 1977). As a whole there exists only a small bulk of data on TF in Callitrichidae and no information on the question whether we are right in assuming that siblings are even- ranking up to TF. EnGEL (1986) has shown that twins already reveal remarkable differ- ences as to their social competence in early infancy. We do not know as yet however which event(s) or process(es) may be responsible for these differences in the relative position of the twins ın the group hierarchy. This would mean that TF must have other and/or different function(s) as was hitherto believed. Up to now the published data on TF ın marmosets and tamarins are too small for a sufficient explanation of the functional aspect of this behavioural phenomenon. Acknowledgements We are greatly indebted to Miss T. GATEsMaNn for her help in translating this artıcle into Englısh. Zusammenfassung Auftreten und Häufigkeit von Zwillingskämpfen beim Weißbüscheläffchen (Callithrix jacchus) Zwillingskampf wurde in 32 von 39 Gruppen und in 27 von 83 Würfen des Weißbüscheläffchens Callıthrix jacchus beobachtet. 39 Würfe waren heterosexuell, 23 isosexuell weiblich und 16 isosexuell männlich. Am häufigsten waren Zwillingskämpfe in reinen Männchen- (50 %) und isosexuellen Weibchenwürfen (46 %), am seltensten (17 %) unter heterosexuellen Wurfgeschwistern. Das Zwil- lingskampf-Alter betrug bei männlichen Wurfgeschwistern im Durchschnitt 188.5 Tage, bei weibli- chen 241.7 Tage und bei heterosexuellen 237.1 Tage. Zwillingskämpfe waren über den gesamten Intergeburtenabstand des «-Weibchens verteilt, fünf zeigten eine zeitliche Beziehung zum 1. bzw. 2. postpartum-Ostrus der Mutter. In 6 % aller Zwillingskämpfe mußte ein beteiligtes Tier aus der Gruppe entfernt werden, um schwere Verletzungen zu vermeiden. Zwillingskämpfe konnten nur in Gruppen mit max. zehn Mitgliedern beobachtet werden, wobei ein relativ hoher Prozentsatz (37 %) ın sehr kleinen Gruppen und in Abwesenheit gegengeschlechtli- cher Gruppenmitglieder (Eltern ausgenommen) zu verzeichnen war. Literature AsBoTT, D. H. (1978): Hormones and behaviour during puberty in the marmoset. In: Recent Advances in Primatology. Ed. by D. J. CHivers and ]. HERBERT. London: Academic Press. Vol. 1, 497-499. ABBOTT, D. H.; HEArRn, J. P. (1978): Physical, hormonal and behavioural aspects of sexual development in the marmoset monkey, Callithrix jacchus. J. Reprod. Fertil. 53, 155-163. EnGeL, C. (1986): Observations on the interaction between adult infant-carrying anımals and group members without rearing experience in the common marmoset, Callıthrix jacchus. Folia primatol. 45, 225-235. EpPLE, G. (1975): The behaviour of marmoset monkeys (Callithrichidae). In: Primate Behaviour. Ed. by L. A. RosengLum. New York: Academic Press. Vol. 4, 195-239. FAGEnN, R. (1981): Play Behavior. New York: Oxford University Press. KLEIMAN, D. G. (1977): Monogamy in mammals. Q. Rev. Biol. 52, 1-67. — (1979): Parent-offspring-conflict and sibling competition in a monogamous primate. Am. Nat. 114, 753-760. Kreıman, D. G.; Mack, D. S. (1977): A peak in sexual activity during mid-pregnancy ın the golden lion tamarın, Leontopithecus rosalıia (Primates, Callitrichidae). J. Mammalogy 58, 657-660. König, A.: Zur Stellung des «-Männchens in Kleingruppen bei Calhthrix jacchus Erzleben, 1777 ın Gefangenschaft unter besonderer Berücksichtigung der Aufzucht eigener und genetisch nicht verwandter Jungtiere (in prep.). ROTHE, H. (1975): Some aspects of sexuality and reproduction in groups of captive marmosets (Callithrix jacchus). Z. Tierpsychol. 37, 255-273. 332 H. Rothe, A. König, Ute Radespiel, K. Darms und Margaretha Sieß — (1979): Das Ethogramm von Callıthrix jacchus Erxleben, 1777 (Primates, Ceboidea, Callitrichi- dae). Eine morphaktische Analyse des Verhaltens mit besonderer Berücksichtigung des sozialen Umfeldes. Habıl.-Schrift, Göttingen. STEVENSON, M. F.; PooL£, T. B. (1976): An ethogram of the common marmoset (Callıthrix jacchns jacchus): general behavioural repertoire. Anım. Behav. 24, 428-451. SUTCLIFFE, A. G.; PooL£, T. B. (1984): Intragroup agonistic behaviour in captive groups of the common marmoset (Callıthrix jacchus jacchus). Int. J. Primatol. 5, 473-489. Authors’ address: Prof. Dr. HARTMUT ROTHE, ANDREAS KÖNIG, UTE RADESsPIEL, Dipl.-Biol. Kurr DARrMS, MARGARETHA SIEsS, Institut für Anthropologie, Universität Göttingen, Bürgerstr. 50, D-3400 Göttingen, FRG Z. Säugetierkunde 53 (1988) 333-340 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Homing behavior of Wood mice (Apodemus) in a geomagnetic anomaly By J. BoveEt, M. Dorıvo, C. GEORGE and A. GOGNIAT Departement de biologie, Universite Laval, Quebec, Canada, and Institut de physiologie, Universite de Lausanne, Lausanne, Switzerland Receipt of Ms. 1. 6. 1987 Abstract Experiments have been performed to test the hypothesis that a previously described local directional bias in the homing behavior of wild mice resulted from an interference between homeward orientation and a tendency to go up the gradient of magnetic intensity at the site of a weak geomagnetic anomaly. An analysıs of homing success of 440 European woodmice (Apodemus flavicollis and A. sylvaticns) tested at any of four sites affected by the same magnetic anomaly provides only limited support to the hypothesis. The results point to the possibility that a stronger effect of the anomaly could be a drop in homing success in the immediate surroundings of the magnetic trough of the anomaly. Introduction In the early 1960’s, one of us carried out a homing experiment near Lausanne, Switzerland, with European wood mice (Apodemus sylvaticus and A. flavicollis), in which anımals displaced from different cardınal directions had different homing performances (BOVET 1962). Using the terminology of the time, he described the differential pattern as a directional effect (= Richtungseftekt) (e.g. KRAMER et al. 1956, 1958; PRATT and WALL- RAFF 1958). Subsequently, we learned from a paper by MEYER DE STADELHOFEN et al. (1973) that BovEr’s (1962) research area was located in the Jorat magnetic anomaly (see Fig. 1,A). In the mean time, evidence had been accumulating on the role of magnetic cues in bird orientation (reviewed by KEEToN 1974). This led us to perform, in 1975-1977, the series of experiments reported in this paper to evaluate the possibility that the directional effect described by BovET (1962) was linked to the magnetic anomaly, and that natural magnetic cues could aftect the orientation of mice. The results obtained were not clearly interpretable in terms of a magnetic effect at this time and were filed away. Since then, experiments with pigeons have led to a better characterization of how and when their homing orientation and/or performance is affected after release in varıous kinds of magnetic anomalıes (review in WAGNER 1983). On the other hand, several experiments have been performed with rodents, some of which suggest an important role of magnetic cues in the homing orientation of these anımals (MATHER and BAkER 1981; Aucust 1985), others not (ETIENNE et al. 1985, 1986). These recent developments on the role of magnetic cues in the spatial orientation of higher vertebrates prompted the present re-examination of the results of the 1975-1977 experiments. The Jorat magnetic anomaly, a part of which is shown in Fig. 1, comprises a faırly uniform disruption of the normal pattern of gradients of total magnetic field intensity, that affects an area of roughly 50 x 35 km (MEYER DE STADELHOFEN et al. 1973). BovET’s (1962) major basis for assuming a directional effect was that in the area he worked in (around A in Fig. 1), the homing success (= ratio of number of successful homers to number of anımals tested) was highest for anımals which had to travel up to 1 km South in order to home, null for anımals which had to travel toward North, and intermediate U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5306-0333 $ 02.50/0 334 J. Bovet, M. Dolivo, ©. George and A. Gogniat toward East and West, with slightly higher figures toward East than West (Fig. 2). Considering that the magnetic isopleths in the area around A run along an ENE-WSW axis and the gradient goes up toward SSE (Fig. 1), we hypothesized that, in ad- dition to a homeward ten- dency, the mice may have a tendency to go up the abnormally steep magnetic gradient (i.e., about 70 nlI/km vs. about 10 nT/ km in normal conditions, at ground level). The con- vergence of the two ten- dencies should facilitate homing, whereas their di- vergence should impaır it. This would account not only for the strong differ- ence between South and North and the intermedi- ate values for Fast and West, but also for the Fig. 1. Map of the Jorat magnetic anomaly. Numbers on isodynamic small difference between lines indicate total magnetic intensity at ground level (in nT, Fast and West: animals +46000). A, B, C, D are the central points of the four experimental areas. R is point 530,000/160,000 of the Swiss reference grid system. homing toward East go Stripped area is part of Lake of Geneva. (Redrawn from MEYER DE slightly u the gradient; STADELHOFEN et al. 1973, with permission) anımals homing toward West go slightly down the gradient. Our 1975-1977 experiments were then aimed at testing the following predictions ot the hypothesis with respect to homing success: Il; 4. New displacements performed in the area around A should reveal the same directional effect as in 1962. Displacements performed in an area such as that around B (Fig. 1), where the magnetic isopleths have the same orientation and the gradient goes up in the same direction as ın the area around A, should reveal the same directional effect. . Displacements performed in an area such as that around € (Fig. 1) where the magnetic isopleths are oriented NNE-SSW and the gradient goes up toward WNW, should reveal a directional difference such that homing success toward West is high, toward East low, and toward North and South intermediate with a small advantage for North. Displacements performed in an area like that around D (Fig. 1), which ıs in a magnetic trough, should not reveal any directional difference. | The hypothesis and its predictions were based on the 1962 experiment and on the description of the anomaly by MEYER DE STADELHOFEN etal. (1973), with no specific input from other sources. However, during the tests, they received support from a paper by FrEr and WAGNER (1976). These authors reported then that homing pigeons released at various sites within the very same Jorat magnetic anomaly had an initial orientation which was a Homing behavior of Wood mice (Apodemus) in a geomagnetic anomal,y 995 compromise between homeward orientation and a tendency to fly up the magnetic gradient, thus something very similar to our hypothesis. On the other hand, the evidence eventually produced by MATHER (1985) suggests that the magnetic de- posits found in the head of woodmice make them potentally as sensitive to magnetic cues as, for instance, pigeons. According to GouLD (1982), behavioral evidence suggests this sensitivity to be at least 10-30 nT in pigeons; also, the single domains ot magnetite discovered ın pigeons could produce a resolution better than 1 nT. Taken to- gether, these data suggest that, along the steepest slope of the gradient ın the Jorat magnetic anoma- ly, woodmice could detect differences in field strength over distances of the order of 200 m or less. Our hypothesis was, therefore, not un- reasonable in this respect. DIEBE 2 [o) PERCENT HOMERS D (=) lei SEWN SEWN 500 1000 Fig. 2. Results of the 1962 experiment: homing success obtained at various sta- tions around release site A (see Fig. 1), ın percent of anımals that homed. Numbers in boxes indicate numbers of anımals dis- placed. S, E, W, N correspond to direc- tion of home site, as seen from A. 500 and 1000 are distances (in m) from release site to home site. Vertical arrows point to zero values. Difference in homing success between South and North was nearly significant at 500 m, and significant at 1000 m (p = 0.094 and p = 0.019, respec- tively; Fisher’s exact probability test, one-tailed). Concordance between the two distance conditions for a decrease of homing success figures in the order S-E- W-N was nearly significant: Spearman’s rs = 0.95) Material and methods The experiments were performed in four areas centered around points A, B, C, and D shown on Fig. 1. Grid references according to Carte nationale de la Suisse (1:25000) are: A= 543,900/159,800; B= 544,850/ 160,280; C= 554,600/157,300; D = 544,800/165,500. The point of lowest magnetic intensity recorded by MEYER DE STADELHOFEN et al. (1973) within the contour of the 46250 nT isopleth ıs located 1200 m ESE of D (their station 172: 46238 nT),. In each area, a number of trapping stations were established according to a symmetrical pattern around the central point. Each station had 20 home-made single catch live-traps, on 4 rows of 5 traps forming a grid of 10 x 10 m squares. We tried to find areas as large as possible that would be at an appropriate location with respect to geomagnetic characteristics, and would cover homogeneous habiıtat. It ıs extremely difficult to find large areas of homogeneous habitat in a densely populated country like Switzerland, and we had to content ourselves with areas that were essentially “wooded” (with a variety of deciduous and coniferous species), with no major clearıngs. Around A and B, we could place 12 stations each, at 250, 500, and 750 m North, East, South and West of the central point. Around C, we could place only 8 stations, at 250 and 500 m North, East, South, and West of the central point. Around D, we were limited to 6 stations at 250, 500 and 750 m South and North of the central point. In any one area, all the stations were operated sımilarly and simultaneously. Details on trapping operations are shown in the Table. All the traps were checked every morning after a nıght of operation. All the mice were marked at first capture with numbered ear-tags and with toe-clips. Only mice considered as residents of a home range overlapping a single trappıng station were used in the experiments. Criteria for residency were that a mouse should have been captured at least three times on at least five consecutive calendar days, and should have been always captured at the same station. All the mice used in a given area were displaced and released at the central point of that area, at places marked A, B, C, and D in Fig. 1. Displacements were performed by transporting the mice on foot and by car ın individual jars without view of the surroundings. Routes followed during displacement resulted from a compromise between avoiding long detours for the mice and allowing the trapper(s) to complete trap checking between sunrise and noon at the latest. All displacements were followed by release on the same day. Depending on when a given mouse was displaced, traps were available for recapture at the home station a variable number of nights. For the last displaced mouse in any area, traps were available for at least 5 nights after release, spread over at least 7 calendar days. There was no obvious natural barrier between any home site and the corresponding release site. All the occasional rivulets and creeks were of a type passable by mice (Bover 1965). The only two-lane paved road that went between several home sites and release sites (in areas A and B) was of a type passable by mice (KozEL and FLEHARTY 1979) and was actually crossed by several successful homers (which could not have homed without doing so). 336 J. Bovet, M. Dolivo, C. George and A. Gogniat Trapping schedules. (Dates formated D.M.Y.) Area Numberof Begin of First Last End of Number of Number stations trapping displacement displacement trapping nights of of mice trapping displaced 12 16.210.775 10875 ZEN, 7 A275 95 12 953827 Re) 9, 1110678 ul, Io, 78 Si 8 6.4.7.6 46 48 IE 6.716 196976 141° 8 1299777. O7. 610,977 1077, 23107 ZZ OST 277 5 * 1976 and 1977 pooled for areaC. The experiments involved 440 Apodemus, each used only once. Numbers of mice studied in any one area are shown in the Table. Of these, 245 (145 &d and 100 2 ?) were identified as A. flavicolhs, and 174 (98 dd and 76 2 2) as A. sylvaticus. The remaining 21 mice (10 dd and 11 ? 2) could not be attributed safely to either species, due to “intermediate” field identification characters. For the analysis of results, all 440 mice were pooled, irrespective of the species they belonged to. According to MErMmoD (1969), home range sizes are similar in the two species, which, as a rule, implies similar homing performances (see ANDERSON et al. 1977; BovET 1978). Source for statistical tests was SIEGEL (1956). Results Homing success values obtained at any one trapping station are shown in Fig. 3. Due to the well known depressing effect that displacement distance has on homing success (BOVET 1978), comparisons related to our magnetic hypothesis can be meaningfully made only ejeJs]e] [elzlnls] [elelelel EEZIE 80 80 so A 60 C 40 40 20 20 : | Ir SEWN SEWN SEWN WNSE WNSE S 250 500 750 250 500 I 2 o [elefela QEIEIE wu 80 80 a so B 60 D 40 40 20 20 SEWN SEWN SEWN SN SN SN 250 500 750 250 500 750 Fig. 3. Homing success obtained at the various stations in areas A, B, C, and D, in percent of anımals that homed. Numbers in boxes indicate numbers of animals displaced. S, E, W, N correspond to direction of home site, as seen from release site. 250, 500, 750 are distances (in m) from release site to home site. Vertical arrows point to zero values Homing behavior of Wood mice (Apodemus) in a geomagnetic anomaly 337 among stations that are at the same distance from the release site. An analysıs of the relationships between homing success and distance in areas A and B has been published elsewhere (BovEr 1982). Areas around A, B, or C For each displacement distance in any area around A, B, or C, the four histogram columns displayed in Fig. 3 are arranged in descending order based on the predictions of the hypothesis. Of the 48 possible comparısons between any two values obtained in the same area and over the same distance, only three reveal a significant difference at the 0.05 level (x? or Fisher’s exact probability test, one-tailed): area B, 750 m, E>W; area C, 500 m, W>N, and W>E. I, in an attempt to overcome the effect of small sample sızes, we pool the comparable values obtained around A, B, and © (i.e., all results obtained over a given distance in directions corresponding to expected highest, second highest, third highest or lowest levels, respectively), the statistical results remain essentially the same. A significant difference is found in only two cases out of 18: 500 m, expected highest > expected lowest; and 750 m, expected second highest > expected third highest. It is thus obvious that, at this level of analysis, the statistics do not confirm the predictions of the hypothesis. There is however a trend, apparent in Fig. 3, for homing success to be higher in those circumstances where the hypothesis predicts it to be higher. Considering values that correspond to expected highest and lowest levels in any set ot comparable figures (S vs. N around A or B, W vs. E around C), a sign-test shows the former to be significantly more often the highest of the two than expected by chance (N = 7; x = 0; p = 0.008). If we give up the rather stringent prediction of a consistent difference between expected second and third highest levels and we pool these two into a single “intermediate” level, there ıs a significant concordance among the experimental conditions for a decrease of homing success from expected highest through intermediate to lowest levels (Kendall coefficient of concordance; N= 3, k = 8, s = 51.5; p<0.05). Area around D There is no significant difference in homing success between South and North over the three distances tested. Although thıs fits the prediction of the hypothesis, ıt has no real supporting value because the similar analysıs applied to the other areas does not verify the contrasting predictions. Not predicted by the hypothesis, homing success over any one distance was steadily lower around D than around either A, B, or © (all directions pooled). The differences were particularly strong over 500m (D35°C, however, the experiments were disrupted earlier when the anımals became restless. Calculation of gaseous exchange started only after an equilibration time of two hours the reported values representing the means over the following period of ca. 3h. Immediately after opening the respiration chamber T,. and body mass were measured. Skin temperatures (T,) were obtained with a special surface temperature probe (United Systems Corp., model 581C, 709A). During these experiments the cuis remained in their normal cages and T, in the room was set to 15, 22 and 34°C, respectively. Skin temperatures were measured 3 and 7.5 h after changing T.,. To find out about the nycthemeral varıiation of locomotor activity the cuis were watched using an infra-red sensitive video-camera (Grundig). Each of the adult couples was observed twice over a complete 24 h period. Length of active and resting periods was registered and occurrence of certain activities noted (e.g. feeding, drinking, grooming, sand-bathing, defecating). Results Nycthemeral variations of T,. In the females no obvious diurnal rhythm of T,. could be detected (Fig. 1). Mean T,. during L-phase and D-phase were nearly identical: 37.3 £ 0.2°C vs 37.4 # 0.2°C. A sımilar situation was found in the males; however, ın this sex a marked peak occurred around the time when the lights were switched off in the anımal room. Mean T,. then was at 38.4 # 0.2°C but dropped to 37.3 # 0.2°C within the next 2h. T;e after exposure to various ambient temperatures In both sexes T,. was regulated within the normothermic range (36.5-38.5°C) at ambient temperatures from 15-35°C (Fig. 2). Exposure to higher T, led to a rapid rise of T,. (to above 41°C after 50-75 min at T, 40°C). Experiments then usually had to be disrupted because the cuis showed signs of severe heat stress (restlessness, defecation, urination). O,-consumption at T, from 15-40 °C In the females oxygen consumption was markedly lower than in the males over the whole range of tested ambient temperatures (Fig. 3). In both sexes minimal O,-uptake occurred at T, from 30-35 °C. Within this thermoneutral zone (TNZ) the basal metabolic rate (BMR) of the females was 17 % below and that of the males 15 % above the mass-specific value as expected after the formula of KLEIBER (1961). At T, below 30°C O,-consumption increased linearly in the females whereas in the males it showed a more irregular pattern. The increase can be described by the following regression equations: Activity pattern and thermoregulation in Galea musteloides 343 females: Y(ml O,/g-h) = 1.795 - 0.038-T, (°C) (r = -0.91) males: Y(ml O,/g-h) = 2.493 - 0.049- T, (°C) (r = -0.73) The slope of the regression lines is 30 % (females) and 12 % (males), respectively, below the calculated mass-specific values after the formula of HERREID and Kesser (1967). With the exception of T, = 15°C mean values of females and males are significantly different (pe 05). At ambient temperatures above the TNZ O;,-consumption increased only moderately. a TG 39 99: dd 38 37 > 36 39 99 temperature Ww — Rectal & — U L D L Time (hours) Fig. 1. Nycthemeral variation of rectal temperature (mean values + S.D.) in cuis (3 males, 4 females). Each anımal was measured twice at every full hour L2 Tre (°0) 15 20 25 30 35 40 Ta 0 Fig. 2. Rectal temperature after exposure to ambient temperatures from 15-40 °C. Exposure lasted for five hours except for ambient temperatures above 35°C when experiments were disrupted earlier to spare the animals. Empty circles = females, filled circles = males; the line represents the mean value of both sexes 344 E. F. Müller and U. Soppa (ml -g1-h1) 0,-consumption > 0,5 15 20 25 30 35 40 Ta (0) Fig. 3. Oxygen uptake at ambient temperatures from 15-40°C (calculated mean after an equilibration time of ca. 2 hours; empty circles = females, filled circles = males). Lines represent regression lines for the increase of O>-consumption at temperatures below the TNZ: ------ = 22: Y (ml O,/g-h) = 1,795 —- 0,038-T, (°C); r= -0,91; ------ = 6d: Y (ml O,/g-h) = 2,493 — 0,049-T, (°C); r= -0,73; = 2+8: Y (ml O,/g-h) = 2,111 - 0,042-T, (°C); r = -0,61 T,. and T, at different ambient temperatures under normal caging conditions In five cuis (3 temales, 2 males) we measured T,. and T, at various sites of the body (naked spot above sternal region, inner side of earlaps, hind foot pad, tip of the nose) after exposure to ambient temperatures 15, 22 and 34°C, respectively (Fig. 4). During these experiments the anımals were kept in pairs in their normal cages. After 7.5h T,. and T, 40 30 S 20 ü 40 Ta 720 3 = ® er ° 30 c = 0] je 20 o 40 Ta=15°C B B Fig. 4. Skin temperatures at the tip of the nose (N), 30 hind foot pad (F), inner side of earlap (E) and bare patch over sternal region (S) after 3 (left column) and 7 (right column) hours of exposure to ambient tem- peratures 15, 22 and 34°C. R = rectal temperature. 20 Height of columns indicates mean value, line indi- cates $.D. (n = 5; 3 females, 2 males) Activity pattern and thermoregulation in Galea musteloides 345 were not significantly different from those after 3 hours exposure. At all tested ambient temperatures T, was lowest at the tip of the nose. The skin temperature of the sternal region and of the earlaps always remained above 30°C. It fell below this level at the tip ot the nose at T, 22°C and, additionally, at the hind foot pad at T, 15°C suggesting an energy saving reduction of the heatflow to these parts of the. body at lower ambient temperatures. 60 n oXe) z = 38 < = nn 3778 Di Sn 36 5 ‘= m = 60 39, = [o] = dd 5 SW 387 = A = {m < 20 37 a 0 36 L D L Time (hours) Fıg. 5. Nycthemeral varıation of time spent actıve. Mean values calculated for periods of 60 min. The anımals were observed with an infrared video device in their normal cages. Each couple was watched twice for a full 24 h period. Lines represent nycthemeral varıation of rectal temperature (see Fig. 1) _ (©) & dd IS N oo L Al IH ad] ef I = 10 N 2? ® I Dr Eh JH a ee > 2 20 dd + 99 (8) oO © _ 10 ® Be} Fig. 6. Relationship between frequency and dura- * 9 EB hob tion of single activity bursts during D-phase (dot- 2555210, 510.156,51522007,20.255725,25 ted columns) and L-phase (empty columns) En of single activity bursts (min) Activity pattern under normal caging conditions In the males no obvious nycthemeral variation of the activity pattern could be detected (Fig. 5). Periods of increased activity were rather uniformly distributed over L- and D- phase. If at all, longer lastıng periods of increased activity in the male sex occurred before and shortly after switching the lıghts off and then again during the fırst half of the D-phase. The females, on the other hand, showed a different pattern of activity: Two marked peaks occurred at the beginning and at the end of the first half of the D-phase. Whereas the activity level remained fairly high throughout the whole D-phase it was markedly lower 346 E. F. Müller and U. Soppa when the lights were on. During the L-phase both sexes spent about the same time active (17.4 % and 17.5 %, respectively). During the D-phase, however, activity increased to 32.9 % of the total time in the females and to only 18.5 % in the males. Durations of single activity bursts were also somewhat different in both sexes (Fig. 6). Whereas the males preferably were active for periods shorter than 15 min (with a clear maxımum of activity bursts <5 min during the D-phase) the females showed a much stronger tendency for extended periods of actıvity (duration >15 min). Visual observation revealed that almost any activity burst was accompanied by the intake of food. Drinking, on the other hand, was almost never seen. Discussion The most striking results of our investigations certainly are the marked sexual differences in the level of heat production and in the pattern of daily activity. As yet it ıs unclear why the basal metabolıc rate of female cuis ıs much lower than that of males (0.69 vs 0.94 ml O,/ g-hat T, 30°C). It must be emphasized that this difference in heat production was found at all tested ambient temperatures although rectal temperatures differed only slightly between the sexes. | Sexual differences in heat production have also been reported for a gerbil, Gerbillus perpallidus (MÜLLER 1985). However, in this desert species BMR was lower in the male sex which, additionally, had also a lower rectal temperature. Reliable information about different levels of heat production among the sexes of other species is unfortunately missing. One can, therefore, only speculate if the peculıarities in the energetics of gerbils and cuis are related to their different reproductive strategies. The higher levels of heat production and rectal temperature in the females of the altricial gerbils could allow a relatively faster development of the young ones during the prenatal period. The differences in the energetics of female and male cuis are probably also reflected in their different activity patterns. Our observations revealed that female cuis spend substan- tially more time actıve than males, the increased activity almost entirely occurring during the D-phase. It seems possible that the lower basal heat production in the females is an adaptation to keep their daily energy expenditure at a level similar to that of the males in spite of their increased activity. This view is agaın corroborated by findings in Gerbillus perpallidus where preliminary results suggest a lower level of activity in the females which - in this species - show a higher rate of basal heat production than males (MÜLLER 1985 and unpubl. observ.). Our results indicate a clearly polyphasıc activity pattern of cuis in both sexes. Activity bursts occur distributed over the entire 24 hours period. As stated above, however, the females are almost twice as active during the night than during the day. This finding is contradictory to Roop’s (1972) and WEıR’s (1972) observations. These authors classıfy cuis as being day-active. However, RooD watched his anımals only for two hours in the morning and again in the evening. This limited observation time probably does not allow to obtain a detached view of the entire nycthemeral rhythm. Measuring the composition of the urine, on the other hand, BELLamy and Weır (1971) concluded that cuis should be night-active. From the data available at the moment it seems most probably that cuis are polyphasic with a rather uniformly distributed activity in the males (with a small peak around the transition from L-phase to D-phase) and a clear accentuation of activity during the D-phase in the females. A comparison of our results with those obtained from guinea pigs is rendered difficult because investigations in this species have either only been made with males or the authors do not state the sex of their animals. However, as pointed out above there may exist marked differences between the sexes with regard to activity levels, activity patterns and Activity pattern and thermoregulation in Galea musteloides 347 heat production. Further studies in guinea pigs and related species dealing with these subjects should carefully watch this fact. Even a comparison of male cuis with male guinea pigs is difficult because there is considerable discrepancy among authors concerning the nycthemeral rhythm of cavies. From their observations of anımals living under natural or near-natural environmental conditions Fuchs (1980), Kıng (1956), and PEArson (1951) concluded that guinea pigs are predominantly day-actıve. Also Kayser and HırpweEin (1974) reported a tendency towards day-time activity ın caged anımals. NıcHoLLs (1926) found an almost uniform distribution of the locomotor activity during both D- and L-phase. Her findings were supported by measurements of the sleep-wake states which exhibited a polyphasic pattern throughout the entire 24 h period (PELLET and BERAUD 1967). On the other hand, STUPFEL et al. (1981), measuring locomotor activity and CO,-production, found evidence that guinea pigs are significantly more active during the night-time. BÜTTNER and WOLLNIK (1982) obtained sımilar results and, additionally, found a well-marked peak of activity around the L:D-transition. In a recent paper JıLGE (1985) reported peaks of activity around the L:D- and D: L-transition with a less pronounced peak around the mid of the D-phase. In his experiments activity was generally somewhat higher during the L-phase. It remains unclear if these contradictory findings occur due to a broad genetic varıability of rhythmic components in the guinea pig as suggested by BÜTTNER and WoLLnIk (1984) or must be ascribed to different methodological approaches used by the various authors. In view of this situation it seems that — regarding the males only - our results in the cuis agree most with those of JILGE (1985) in male guinea pigs. It would be very interesting to see ıf female guinea pigs exhibit the same different activity pattern as has been found in female cuis and if ıt ıs also connected with sımilar peculiarities in their basal energetics. As long as this information is missing it remains difficult to understand the adaptive value of the sexual differences found in cuis. Acknowledgement We are indebted to Dr. D. BÜTTNER, Essen, for critically reviewing an earlier draft of the paper. Zusammenfassung Aktivitätsmuster und Temperaturregulation beim Wieselmeerschweinchen (Galea musteloides Meyen, 1833) Untersucht wurden an sieben Wieselmeerschweinchen (Galea musteloides) basale Wärmebildung, temperaturregulatorische Reaktionen und täglicher Aktivitätsverlauf. Bei weiblichen Wieselmeer- schweinchen lag die mittlere Rektaltemperatur am Tage und in der Nacht bei 37,3-37,4°C. Bei den Männchen zeigte sich dagegen ein deutlicher Gipfel zur Zeit des Überganges von der L- zur D-Phase. Im Bereich von T, 15-35°C wurde T,. zwischen 36,5-38,5°C reguliert. Aufenthalt bei höheren Temperaturen führte zu einem raschen Anstieg von T,.. Der Sauerstoffverbrauch war am niedrigsten zwischen 30-35°C; der Basalstoffwechsel der Weibchen lag dabei sehr viel niedriger als bei den Männchen (0,69 gegenüber 0,94 ml/g-h). Diese Werte liegen 17 % unter (Weibchen) bzw. 15 % über (Männchen) den gewichtsspezifischen Erwartungswerten. Auch die Zunahme des O,-Verbrauchs bei Temperaturen unterhalb der TNZ verlief bei den Weibchen deutlich flacher: Die Steigung der Regressionsgeraden war 30 % (Weibchen) bzw. 12 % (Männchen) geringer als der errechnete gewichtsspezifische Wert für die Wärmedurchgangszahl. Messungen der Hauttemperatur an verschie- denen Körperteilen ergaben, daß in kalter Umgebung der Wärmetluß zur Peripherie hauptsächlich im Bereich der Nasenspitze und der Fußsohlen reduziert wird. Unter normalen Haltungsbedingungen unterschieden sich die Aktivitätsmuster der Geschlechter deutlich. Während bei den Männchen die Aktivität ziemlich gleichmäßig über D- und L-Phase verteilt war, zeigten die Weibchen während der Nacht fast eine Verdoppelung der Aktivität. Außerdem war bei ihnen eine stärkere Tendenz zu länger andauernden Aktivitätsschüben zu erkennen. 348 E. F. Müller and U. Soppa References BerrLamy, D.; WEIR, B. J. (1972): Urine composition of some hystricomorph rodents confined to metabolism cages. Comp. Biochem. Physiol. 42A, 759-771. BÜTTNER, D.; WOLLNIK, F. (1982): Untersuchungen zur Kurzzeitperiodik beim Meerschweinchen (Cavia aperea f. porcellus). 1. Bewegungsaktivität unter Licht-Dunkelwechsel, Dauerdunkel und Dauerlicht. Z. Säugetierkunde 47, 370-380. BÜTTNER, D.; WOLLNIK, F. (1984): Rhythmic components of locomotor activity and food intake in the guinea pig. In: Standards in laboratory anımal management. UFAW, Hamilton Close. pp. 189-196. CABRERA, A. (1961): Catalogo de los Mammitferos de America del Sur. Buenos Aires: Revta. Mus. argent. Cienc. nat. Bernardino Rivadaviıa. Fuchs, $. (1980): Spacing patterns in a colony of guinea pigs: predictability from environmental and social factors. Behav. Ecol. Sociobiol. 6, 265-276. HERREID, C. F.; Kesser, B. (1967): Thermal conductance ın birds and mammals. Comp. Biochem. Physiol. 21, 405—414. HückıncHaus, F. (1961): Zur Nomenklatur und Abstammung des Hausmeerschweinchens. Z. Säugetierkunde 26, 108-111. JILGE, B. (1985): The rhythm of food and water ingestion, faeces excretion and locomotor activity in the guinea pig. Z. Versuchstierkunde 27, 215-225. Kayser, C.; HıLpwein, G. (1974): Evolution de la consommation d’oxyg£ene et de l’activite du cobaye au cours du nyctheme£re. Arch. Scı. Physiol. 28, 1-23. King, J. A. (1956): Social relationships of the domestic guinea pig living under semi-natural conditions. Ecology 37, 221-228. KLEIBER, M. (1961): The Fire of Life. New York: Wiley. MÜLLER, E. F. (1985): Untersuchungen zur Temperaturregulation bei der Wüstenrennmaus Gerbillus perpallidus Setzer, 1958. Z. Säugetierkunde 50, 337-347. NicH#outs, E. E. (1922): A study of the spontaneous activity of the guinea pig. Comparative Psychology 2, 303-330. Pearson, O. P. (1951): Mammals in the highlands of southern Peru. Bull. Mus. Comp. Zool. 106, 117-174. PELLET, J.; BERAUD, G. (1967): Organisation nycthemerale de la veille et du sommeil chez le cobaye (Cavia porcellus). Comparaison interspecifiques avec le rat et le chat. Physiol. Behav. 2, 131-137. Roop, ]J. P. (1972): Ecological and behavioural comparisons of three genera of argentine cavıes. Anım. Behav. Monogr. 5, 1-83. RowLanos, I. W.; WEIR, B. J. (1974): The Biology of Hystricomorph Rodents. Symp. Zool. Soc. London 34. London: Academic Press. STUPFEL, M.; PERRAMON, A.; GOURLET, V.; THIERRY, H.; LEMERCERRE, C.; DAVERGNE, M.; Monvoıisin, J. L.; DA SıLva, J. (1981): Light-dark and societal synchronization of respiratory and motor activities in laboratory mice, rats, guinea-pigs and quails. Comp. Biochem. Physiol. 70A, 265-274. Tam, W. H. (1973): Progesteron levels during the oestrus cycle and pregnancy in the cuis, Galea musteloides. J. Reprod. Fert. 35, 105-114. WEIR, B. J. (1971): The evocation of oestrus in the cuis, Galea musteloides. J. Reprod. Fert. 26, 405408. WEIR, B. J. (1972): Laboratory hystricomorph rodents other than the guinea-pig and the chinchilla. In: UFAW Handbook on the care and management of laboratory anımals. 4th ed. UFAW, Edinburgh and London: Churchill Livingstone. pp. 278-286. Authors’ address: Dr. E. MÜLLER and U. Soppa, Institut Biologie III der Universität Tübingen, Abt. Physiologische Okologie, Auf der Morgenstelle 28, D-7400 Tübingen, FRG Z. Säugetierkunde 53 (1988) 349-357 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Die Bedeutung des Untergrundes für die Größe von Bauen des Dachses (Meles meles) am Beispiel zweier Gebiete Südostbayerns Von W. F. Bock Institut für Haustierkunde, Universität Kiel Eingang des Ms. 23. 6. 1987 Abstract The importance of soil composition for the size of badger setts (Meles meles), in two areas in SE Germany The aim of this study was to analyse the soil composition at the sites where badgers dig their setts and to determine which soil composition they prefer. 148 sets ın two areas, the Bavarıan Alps and in the alpine foot hills were analysed. A subjective 4 category scale was used to describe the ease of digging the soil. Setts in the two areas were distributed differently over the categories and in setts in rocks and setts in soil. Most badger setts were dug in soils containing nearly 10% clay, 15% sılt and 75 % fine sands. Sandy soils were favoured by badgers (4? = 36.86, p <0.001) and here the most extensive setts were found (H-Test, p <0.001). Einleitung Dachse sind nicht gleichmäßig über ihre Habitate verbreitet, dies mag u.a. auch an der Eignung der Böden zum Graben von Bauen liegen. Die Meinungen über die von Dachsen bevorzugten Bodenarten sind nicht einhellig. Viele Autoren gehen auf den ökologisch bedeutsamen Faktor Boden nicht oder nur am Rande ein. Die Verwendung von regional eng begrenzt vorkommenden Bodenbezeichnungen erschwert darüber hinaus eine Ver- gleichbarkeit in größerem Rahmen. Nach StusgBE (1982) bevorzugen Dachse tiefgründige, schwere Böden. Near (1977) nennt vor allem sandige Böden als bevorzugtes Substrat für die Bauanlage, weil dieses gut dräniert und leicht zu graben ist. Zu ähnlichen Aussagen kommen auch Dunwer und KiLLinGLey (1969) sowie LIKHACHEV (1956), der zeigte, daß die Kessel, wenn möglich, in sandigem Substrat angelegt werden. Near (1977) berichtet ferner, Baue in Sandböden wären vorhersagbar viel größer als Baue in anderen Böden. Ein objektives Kriterium, nach welchem die Baugrößen festgestellt wurden, teilt er indes nıcht mit. In der vorliegenden Studie wird versucht zu belegen, daß die voneinander abweichen- den Baugrößen zweier getrennter Untersuchungsgebiete auf die unterschiedliche Eignung der Böden zur Bauanlage zurückzuführen sind. Untersuchungsgebiete Die in Südostbayern gelegenen Untersuchungsgebiete sind das Salzachhügelland - ein Teil des Alpenvorlandes — und der Raum des Berchtesgadener/Ramsauer-Beckens, zu den nördlichen Kalkal- pen gehörend. Die südlichen und östlichen Teile des Berchtesgadener/Ramsauer-Beckens liegen z.T. ım Nationalpark Berchtesgaden, dessen Nordrand einen Schwerpunkt des Untersuchungsgebietes darstellte. Würmeiszeitliches Moränenmaterial, von Grobschottern bis zu Feinsanden mit unterschiedlichen Beimengungen von Ton und Schluffen, bildet die vorherrschenden Böden des Alpenvorlandes. Die Wälder sind auf die für eine landwirtschaftliche Nutzung wenig günstigen Moränenhügel, die U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5306-0349 $ 02.50/0 350 W. F. Bock Flußauen oder vermoorte Niederungen reduziert. Grünlandwirtschaft und Ackerbau sind die vor- herrschenden landwirtschaftlichen Nutzungsformen. Das Berchtesgadener/Ramsauer-Becken ist durch relativ weite und reich strukturierte Talzonen charakterisiert. Flachgründige, wenig fruchtbare Böden sind vorherrschend, sie erlauben nur eine Mäh- und Weidenutzung. Der Waldanteil ist mit ca. 50-60 % ziemlich hoch. Material und Methode Es wurden 148 Dachsbaue (Haupt- und Nebenbaue, davon 73 Erdbaue, 61 Felsbaue und 14 Baue unter Hütten, Ablaufröhren etc.) untersucht und ihre Größe ermittelt. Als sehr brauchbar erwies sich dabei das Bestimmen ihres Rauminhaltes unter Zuhilfenahme der davor liegenden Frdauswurfhaufen und einer hierauf aufbauenden Einteilung in Größenklassen, die auch eine Einteilung in Haupt- und Nebenbaue erlaubt (vgl. Bock 1986). Nur die Naturbaue (= Erd- und Felsbaue) werden im weiteren behandelt, da Baue in anthropogenen Strukturen für die angewendeten Beurteilungskriterien keine brauchbaren Daten liefern. Mit einer Grabgabel wurden die Erdauswurfhaufen zum Sammeln von Knochen durchwühlt und die Böden dabei auf ihre Grabbarkeit hin beurteilt. Diese rein subjektiv bewertenden Befunde wurden in vier Grabbarkeitsklassen eingeteilt: Klasse 1 = schlecht bzw. nicht grabbar, 2 = mäßig, 3 = gut und 4 = sehr gut. In diese Grabbarkeitsbewertung sind auch die Felsbaue einbezogen. Von 30 repräsentativen Bauen (8 im Gebirge, 22 im Alpenvorland) wurden Bodenproben aus den Erdauswurfhaufen entnommen und auf Bodenart, Körnung und ihre Eigenschaften (= Bonitierung) und ihre Grabbarkeit hin untersucht. Diese Proben erlaubten durch Vergleich die Zuordnung der restlichen Bodenproben. Im Hinblick auf ihre Grabbarkeit wurden die Werte der Bodenanalyse nur zum Teil berücksichtigt, da z.B. grobe Gesteinsbrocken, die einem Dachs das Graben schwer machen, in den Analysewerten nicht zum Ausdruck kommen. Tabelle 1. Grabbarkeit des Untergrundes in beiden Untersuchungsgebieten sowie getrennt nach Fels- und Erdbauen Grabbarkeitsklasse 4 (sehr gut) 2 (mäßig) 1 (schlecht) Gesamt n 38 46 Gebirge Felsbau % 44 Erdbau % h 47.4 Alpenvorland Felsbau % Erdbau % ä 9.1 Ergebnisse Die Baugrößen Die Baugrößenunterschiede der beiden Untersuchungsgebiete sind drastisch. Die durch- schnittliche Größe der Alpenvorlandbaue (n = 65) liegt bei 12 m? Rauminhalt, die der Gebirgsbaue (n = 69) bei 0,5 m? (Mann-Whitney U Test, p <0,001). Die Grabbarkeit der Böden Nur selten war es möglich, die Grabbarkeit des Bodens am gewachsenen Untergrund zu bestimmen. Deshalb wurde das aus dem Bau gescharrte Bodenmaterial der Naturbaue getrennt nach den beiden Untersuchungsgebieten, in vier Grabbarkeitsklassen eingeteilt, sowie nach Fels- und Erdbauen getrennt (vgl. Tab. 1). In ihrer Grabbarkeit unterscheiden sich die Baue beider Gebiete deutlich (x? = 69,507, p <0,001). Diese Grabbarkeitsbestimmung der Böden vor Ort wurde durch die späteren Boden- analysen ergänzt. Die Bedeutung des Untergrundes für die Größe von Bauen des Dachses 95 Die Bodenanalysen des Erdauswurfmaterials Im Gebirge waren Bodenanalysen nur bei einigen Erdbauen (n = 8) sinnvoll, da bei der großen Zahl von Felsbauen nur das Material der Spaltenfüllungen zu bestimmen ist, welches aber nicht mit dem tatsächlichen Material der Baue identisch ist. Mitunter kann es auch bei Erdbauen vorkommen, daß das von den Dachsen herausgegrabene Material dem Beobachter nur in Form dieser Erdauswurfhaufen und nicht als anstehendes Material Tabelle 2. Analyse von 30 Bodenproben aus Dachsbauen A = Alpenvorland, B = Gebirge Körnung (%) Bonitierung! c d onau Runen mowmwmwumuıu ou$svrane mu wwuWmWw merwWwm Ku U mWNWN WR W u wwuıo nun VUN En a Rum Vv Vu uU RUN RL Dans OS OS ON oowb-+ror-ro-> U WDPWFPEDR-L [© 2 © „ Bere SH SL 7 ui SS 7 er nu pr oo ROO--wuo RR LRND PuvurvVuuen DDP RUN RWLR DHrNDWWRND w DELL RDF 5 0 5 0 5 5 5 0 Die Zahlen in den Klammern geben die Bodengüteklassen wieder, hierbei wurde auch größeres Gesteinsmaterial berücksichtigt, das in der Bodenanalyse nicht aufscheint. Körnung = % Anteile der verschiedenen Bodenarten in einem Bodengemisch; T = Ton, U = Schluff, S = Sand, L = Lehm, H = Humus, # TUS = Ton-, Schluff- und Sandgehalt; BA$ = Bodenart nach Bodenkundlı- cher Kartieranleitung, 3. Aufl. (1983); 1’ = schwach lehmig, | = stark lehmig, t’ = schwach tonıg, t= stark tonig; u’ = schwach schluffig, u = stark schluffig; !a = Carbonatgehalt, b und c = Standfestigkeit des Bodens in feuchtem bzw. trockenem Zustand, d und e = Grabbarkeit des Bodens in feuchtem bzw. trockenem Zustand, f = Wasserhaltefähigkeit des Bodens; 0 = nıcht, 1 = sehr gering/schlecht, 2 = gering/schlecht, 3 = mäßig, 4 = gut/hoch, 5 = sehr gut/hoch; & = die ın 2. Zeile stehenden Werte gelten für toniges Material; * = in fast wassergesättigtem Zustand zerfließend. 352 W. F. Bock zugänglich ist. In jedem Falle ist daher bei diesem Bautyp die Bestimmung der Böden durch Analyse der Erdauswurfhaufen aussagekräftig. Im Alpenvorland wurde der Erdauswurf von 22 Dachsbauen analysiert. Die Ergebnisse der Bodenanalyse aus beiden Untersuchungsgebieten sind ın Tab. 2 aufgeführt. Die größten Bauanlagen des Alpenvorlandes liegen alle in lehmigem Sand. Bei solcher Bodenzusammensetzung sticht für gewöhnlich auch die Größe der Auswurfhaufen ins Auge. Die Grabbarkeit dieser Böden ist sehr gut, wenn sie nicht zu skelettreich sind (als skelettreich werden Böden bezeichnet, die viele anorganische Teilchen >2 mm enthalten). - Sie trocknen rasch ab, erwärmen sich schnell und gewährleisten eine geringe Restfeuchte ım Innern der Baue. Eindringendes Wasser versickert ım allgemeinen schnell. Ist der Tonanteil dieser Sandböden sehr gering, dann hat dies auch nachteilige Folgen. Die geringe Wasserhaltefähigkeit dieser Böden (s. z.B. Tab.2 Nr. 9) bedingt ein sehr schnelles Durchtropfen des Niederschlagswassers. Die Standfestigkeit dieser Böden läßt rasch nach und ihre Fließgrenze wird alsbald erreicht. Zu großes Austrocknen kann andererseits zum Zerrieseln solcher Böden führen. Böden mit einem geringen Sandgehalt, hohen Ton- und Schluffanteilen werden eben- falls von den Dachsen bewohnt. Die hohe Wasserhaltefähigkeit dieser Böden hat zur Folge, daß sie viel Niederschlagswasser aufnehmen können, bevor Wasser durchtropft oder ihre Standfestigkeit merklich nachläßt. Auch bleibt, je nach Skelettanteil des Bodens, eine gute Grabbarkeit über einen langen Zeitraum erhalten. Austrocknung kann bei bindigen Böden das Graben in den äußeren Bodenschichten sehr erschweren bzw. sogar ganz verhindern. Gleichzeitig sind diese Bö- den relatıv kalt, und sie erwärmen sich nur langsam. Da eingedrungenes Wasser nur langsam versickert, bewirkt dies über einen langen Zeitraum eine hohe Restfeuchte in den Bauen. n=38 35 Erdauswurf (m&) 25 Die Beziehung zwischen Bodenmaterial und Baugröße bei Erdbauen Die Böden beider Untersuchungsgebiete wurden in vier Qualitätsstufen eingeteilt (Felsbaue werden gesondert behandelt). Als „schlecht“ (= 1, die niedrigste Bewer- tungsstufe) werden Torfböden und grobe Moränenschotter bezeichnet. „Mäßig“ (= 2) sind Tone und Schluffe mit zum Teil schiefe- rigen Gesteinsbrocken. „Gute“ Böden (= 3) sind skelettreiche Sande aller Art und „sehr gute“ Böden (= 4) sind Feinsande mit gerin- 7 gen Ton- und Schluffanteilen. Wie bei allen Bodengüteklassen künstlichen Einteilungen, so gibt es auch hier zwischen den einzelnen Gruppen Über- Die Baugrößen in Abhängigkeit von der Boden- z güte. (Bodengüteklasse 1 s. Text) saıse In der Abbildung sind die Baugrößen zu den vier Bodengüteklassen in Beziehung gesetzt. Daraus ergibt sich klar, daß Dachse lieber in Sandböden als in anderen Böden graben (%° = 36,86, p <0,001) und in diesen Böden die größten Baue anlegen (H-Test, p <0,001). (Da die Bodengüteklasse 1 nur zwei Werte enthält, war es nicht möglich, sie in der Abbildung darzustellen.) Die Bedeutung des Untergrundes für die Größe von Bauen des Dachses 333 In der stark verinselten Landschaft des Alpenvorlandes fällt die beschränkte Siedlungs- möglichkeit für Dachse auf. Selbst in Wäldern sind Dachsbaue oftmals mangels geeigneter Böden nur an ganz bestimmten Stellen zu finden, meist handelt es sich hierbei um Sandböden würmeiszeitlicher Jungmoränen. Es gibt Gebiete, in welchen sich ein großer ganzjährıg bewohnter Hauptbau befindet, beständige Nebenbaue sind aber nirgends ım Streifgebiet dieser Dachse zu finden. In trockenen Zeiten werden dann Ablaufrohre bezogen und wenn der Mais oder das Getreide eine bestimmte Höhe erreicht haben, legen die Dachse alljährlich an den selben Stellen im Feld Baue an, die genau so regelmäßig von den Bauern wieder zerstört werden. Im Untersuchungsgebiet sind bisher drei solcher Stellen im Feld bekannt. Zwei Baue wurden in Moorgebieten untersucht. In beiden Fällen steht der Grundwas- serspiegel sehr hoch, so daß zwischen Erdoberfläche und Grundwasserspiegel für die Bauanlage höchstens eine 90 cm tiefe Bodenschicht bleibt. Beide müssen wegen der unzureichenden Boden- und Standortverhätlnisse, der hohen Wasserhaltefähigkeit des Bodens und auch von ihrem Raumangebot her zu den schlechten Bauen gerechnet werden. Die spezielle Situation der Felsbaue Viele Felsbaue verfügen nur über ein begrenztes Hohlraumangebot, weshalb in diesen eine multifunktionale Nutzung (Überwinterung, Jungenaufzucht, Bewohnen verschiedener Bauteile je nach Bedarf), wie in „guten“ Erdbauen nicht möglich ist (vgl. Bock 1986). Die Felsbaue sind in drei Typen einzuordnen: Felsblöcke, Kalksteinfelswände und Konglome- rattelswände bzw. -blöcke. Als Felsblöcke werden freistehende bzw. -liegende Felsen bezeichnet, deren Begrenzungen gut erkennbar sind. Ist die Längen- oder Breitenausdeh- nung eines größeren Blockes nicht erkennbar, weil sich ein Teil davon in den Hang fortsetzt, so wird er als Felswand eingestuft. Im Alpenvorland sind Konglomeratfelswände die alleinig vorkommenden Felsbautypen (n= 10), sie nehmen nur 15,4 % der hier gefundenen Naturbaue ein. Felsblockbaue sind im Gebirge eine häufige Bauvarıante (n = 30). Nur unter zweien wurden im Beobachtungszeitraum Junge nachgewiesen. Beide Felsbaue heben sich mit drei weiteren in ihren Ausmaßen deutlich von der Anzahl der übrigen Baue ab (Seitenlänge 10 bis 15 m). Ein Überwintern mit Jungenaufzucht dürfte nur unter diesen großen Felsblök- ken möglich sein. Kleine Blöcke (n = 25) haben im Durchschnitt folgende Ausmaße: L=5 m, B= 3,4 m, H= 3 m. Der Hohlraum, welchen sich die Dachse im günstigsten Falle darunter anlegen können, ist in allen Fällen nur zum Übertagen im Sommerhalbjahr geeignet. Nur dreimal waren ım Winter unter derartigen Blöcken Tiere für höchstens einen Tag festzustellen. Bei einer Anzahl kleiner Blöcke, unter denen telemetrierte Dachse im Sommerhalbjahr übertagten, war deren Reaktion auf den sich annähernden Untersucher gut zu vernehmen. In einem Fall war sogar der abwartende Dachs, der aus dem Lager an den entferntesten Punkt seiner Behausung geflüchtet war, durch das Eingangsloch zu beobachten. Felswandbaue (n = 29), deren größtmögliche Hohlraumdimensionen sich der Beurtei- lung eines Untersuchers völlig entziehen, können in ihren Baumaßen ausschließlich näherungsweise über die Erdauswurfhaufengrößen beurteilt werden (£= 3,5 m°). Ver- gleicht man die Größen von Felswand-Hauptbauen (n = 18) und Erd-Hauptbauen (n = 40) nach ihren Erdauswurfhaufen, so liegen die Felswand-Hauptbaue im Mittel (&) bei 6,2 m’, und sind somit deutlich kleiner als die Erd-Hauptbaue mit 21,4 m? (Mann-Whitney U Test, p < 0,001). Von den sieben Konglomeratfelswänden im Gebirge sind sechs Hauptbaue, die von ihren Hohlraummaßen her mit zu den besten Bauen gehören. Doch sind hierunter auch drei sehr feuchte Baue, von denen zumindest einer zur kalten Jahreszeit in Teilen 354 W. F. Bock unbewohnbar ist. Das durch die Decken troptende Wasser gefriert in den Eingängen zu Eissäulen, wodurch diese für lange Zeit unpassierbar sind. Die ursprünglichen Bodenmaterialien, die aus diesen Bauen gescharrt wurden, sind häufig infolge des darin enthaltenen Humusanteils, der vom Nestmaterial herrührt, nicht mehr mit Bestimmtheit zu ermitteln. Außerdem gilt es hier in erster Linie den Fels zu beurteilen, da es sich bei dem Bodenmaterial in vielen Fällen um Spaltenfüllungsmaterial handelt. Bei vielen Felsbauen - insbesondere bei Konglomeratfelsen — bestehen die zeitlich jüngeren Grabmaterialien, neben dem verrotteten Nestmaterial, fast nur aus grobem Geröll, welches bis zu kindskoptgroße Gesteinsbrocken enthält. Bei den Konglomeratfelsbauen des Alpenvorlandes besteht das Erdauswurfmaterial zum überwiegenden Teil aus schluffig-tonigem Lehm sowie dem humosen Material der zersetzten Nestlaubstreu. Diese Böden sind wegen der in ihnen enthaltenen Dauerfeuch- tigkeit nur bewohnbar, weıl sie durch die Felsplatte abgedeckt sind. Dennoch ist die Mehrzahl dieser Baue während des Sommerhalbjahres kaum bewohnt. Einen Übergang vom Erd- zum Felsbau stellen solche Baue dar, deren Hohlräume blockgestützt sind. Größere sich ineinander verkeilende Blöcke, wie sie z.B. als Folge von Bergstürzen entstehen, bilden Hohlräume in einem sonst völlig unbewohnbaren Boden- material. Diskussion In beiden Untersuchungsgebieten sind -— von Felsdecken einmal abgesehen - stabile, wasserundurchlässige Deckschichten bei Dachsbauen selten. Unter einer humosen Boden- schicht von wenigen Zentimetern ım Gebirge, bis zu 30 cm ım Alpenvorland findet sich ım allgemeinen jene Bodenschicht, in welcher auch der Bau angelegt wird. Da die Mehrzahl der Baue am Hang liegt, besteht auch bei Sandböden mit geringen Ton- und Schluffgehal- ten kaum Einsturzgefahr. Ein Großsteil des Regenwassers läuft infolge des Gefälles bereits oberirdisch und ın der Mutterbodenschicht ab. Weiteres Wasser rinnt entlang den Decken der Kessel und Eingangsröhren ab. Das restliche Wasser, das durch den Kessel tropft, kann das Nest bzw. darinliegende Dachse benetzen. Da aber ın allen Bauen die hohe Luftfeuch- tigkeit eine Verdunstung erschwert, muß eindringendes Wasser möglichst schnell wieder aus dem Kessel abgeleitet werden. Diese Möglichkeit ist bei sandıgen Dachsbauen mit und ohne wasserundurchlässiger Deckschicht gegeben. Bei feuchtigkeitsgesättigter Luft ist es in allen Bautypen gleichermaßen schwierig für die Dachse, ein nasses Fell zu trocknen. Die einzige Möglichkeit — abgesehen von Zugluft, die die meisten Tiere meiden - ist das Trocknen von Fell und Nest durch die Körperwärme. In ebenem Gelände, wo schwerere Böden vorherrschen, können die Dachse, abgesehen von einer Eignung des Bodens, meist deshalb nicht siedeln, weil diese Gebiete ausschließ- lich einer landwirtschaftlichen Nutzung vorbehalten sind und die Landschaft hier in aller Regel einförmig flach ist. Hier wurden durch die Flurbereinigung die Verteilungsmuster der Dachsbaue und daraus resultierend die der Dachse so entscheidend verändert, daß eine rein anthropogen bedingte Verteilung vorliegt. Dachse sind keine primären Waldbewoh- ner, sie besiedeln auch Halbwüsten und Steppen (vgl. HEPTNER und NAaumov 1974). Mehrere Baue im Gebirge liegen an waldlosen Abhängen, doch auch um diese stellt sich bald eine nitrophile Pioniervegetation mit Holunder (Sambucus nigra) und Brennessel (Urtica dioica) ein, die durch die Ausscheidungen der Dachse und deren Erdbewegungen gefördert wird. LuGerrt (1985) führt die auf den Wald beschränkte Verteilung der Dachsbaue ın Schleswig-Holstein auf Störung und Verfolgung zurück. Kruuk (1978) zeigt für die Wytham Woods bei Oxford, daß die dortigen Dachse ihre Hauptbaue zwar alle ın sandigen Böden anlegen, diese aber zugleich im Wald liegen. Die Bedeutung des Untergrundes für die Größe von Bauen des Dachses 355 Nicht so sehr direkte Verfolgung, als vielmehr die Zerstörung eines Teils seiner Lebensräume ist für das Siedeln des Dachses im Wald verantwortlich zu machen. Interessant sind die detaillierten Angaben LikHACHEV’s (1956) über die verschiedenen Bodenprofile bei den von ıhm ausgegrabenen Dachsbauen südlich von Moskau. Lehme sind bis in ca. 130 cm Tiefe die hier vorherrschenden Böden. Sie machen für die Dachse ein Vordringen in die tieferen Bodenschichten recht schwierig. An Abhängen, wo die Boden- schichten durch Gewässerläufe angeschnitten sind, können die Dachse genau jene tieferlie- genden sandigen Bodenschichten angraben, die zur Bauanlage am geeignetsten sind. LIKHACHEV’s Schlußfolgerungen, die Baue würden unter dem Grundwasserniveau liegen, dürften nur bei Schichtböden mit verschiedenen Grundwasserniveaus zutreffen. Früher oder später würden derartige Bauanlagen bei wechselnden Grundwasserständen überflutet werden, wenn die Dachse versuchen, die darüberliegenden Decken zu durchstoßen. Periodisch wiederkehrende Staunässe wäre hier eher vorstellbar. In einer Reihe von Dachsarbeiten aus England werden Angaben zu den Böden gemacht, in welchen die Baue gegraben wurden. DunwELL und KırLinGLeyY (1969) sowie Near (1972) halten eine wasserundurchlässige Deckschicht und eine darunterliegende Sandlage am besten zum Anlegen von Bauen geeignet. Von der großen Anzahl an Dachsbauen, die Harrıs (1984) im Bereich der Stadt Bristol bearbeitete, lag der überwiegende Anteil ebenfalls in Sandstein bzw. sandıgem Boden. CLEMENTS (1974) zeigt, dafs die Dachse, wenn sie die Wahl haben, immer Sandböden bevorzugen. Aus den von GOETHE (1955) vorgestellten Untersuchun- gen ist klar zu erkennen, daß es sich ın seinen Untersuchungsgebieten Teutoburger Wald und Lipperland um suboptimale Böden ım obigen Sinne handelt. Die meisten der von ihm registrierten Baue befanden sich in Lehm und Keupermergelböden. Die Annahme, daß die Mehrzahl der dort gefundenen Baue suboptimal ist, wird durch die für diese Gebiete genannten Nebenbaue in Feldern - hier „Feldnotbaue“ genannt - erhärtet. Bei Felsbauen ist das ausgegrabene Bodenmaterial für eine Baubeurteilung ın anderer Hinsicht interessant. Analysıert man hier die Auswurfhaufen, so findet sich in der unteren Schicht jenes Material, aus welchem die ursprüngliche Spaltenfüllung des Felsgesteins bestand. Darüber liegen humose Erdschichten, die in ihren obersten Lagen zum Teil aus nicht verrottetem Pflanzenmaterial bestehen. Dies ist insbesondere dann der Fall, wenn die Nesteinstreu vorwiegend aus Buchenlaub zusammengetragen wurde. Auf eın in vielen Fällen sehr begrenztes Hohlraumangebot der Felsbaue kann aus zwei Gründen geschlossen werden: Erstens sind hier über die Jahre keine oder nur sehr geringe Grabaktivitäten zu bemerken und zweitens sınd vor diesen Bauen häufiger Skelettreste von verendeten Dachsen zu finden. Die Gesteinsbrocken, die im Gebirge vor allem ım Auswurfmaterial vor Konglomeratfelsbauen zu finden sind, zeigen, daß hier das Rauman- gebot voll ausgeschöpft wird. Wohnungsnot bei Dachsen kommt aber auch in anderen Untersuchungen zum Ausdruck. WIJNGAARDEN und VAN DE PEPPEL (1964) berichten von bis zu 20 cm großen Gesteinsbrocken, die im Auswurfmaterial der Dachsbaue zu finden sind. Ebenso belegen die bei Near (1977) abgebildeten Kalksteinbrocken, auf welchen tiefe, von Dachskrallen herrührende Rillen zu sehen sind, die auch andernorts vorhandene Wohnraumnot der Dachse ın felsigem Gelände. Dort wo Baue kaum anzulegen sind, findet man Dachse auch unter Hütten und in nicht wasserführenden Ablaufröhren. Dies ist besonders häufig im Gebirge der Fall, tritt aber auch im Alpenvorland auf, wo die Habitate der Dachse durch strukturarme Agrarland- schaften eingeengt worden sind. Vor Erdbauen in „guten Böden“ findet man relativ selten die bereits oben erwähnten Überreste verstorbener Dachse. Diese Baue gestatten es den Dachsen, bestimmte Bauteile für einen längeren Zeitraum nicht zu bewohnen, in andere Bauteile auszuweichen oder neue zu graben. Im Gebirge oder in anderen Landschaften, die den Dachsen nur wenig Ausweichmöglichkeiten bieten, ist die Verfügbarkeit genügend guter Baue ein popula- tionsbegrenzender Faktor („gute Baue“ sind Baue mit großen inneren Dimensionen und 356 W. F. Bock einer günstigen Bodenzusammensetzung). Die Baue bleiben kaum länger als ein halbes Jahr unbewohnt und Tierleichen müssen wegen Raumnot — meist bald nach der Zerset- zung — entfernt werden. Die Feststellung, die hohe Mortalität der Jungdachse in den ersten Lebenswochen (bis 25%) sei durch das stickige Bauklima verursacht (NEAL und HARRISON 1958; STUBBE 1965), kann nur für solche Baue zutreffen, deren über die Maßen feuchtes Baumilieu durch eine hohe Wasserhaltefähigkeit und mangelhafte Drainage des Bodens bei schlechter Belüftung verursacht wird. In erster Linie sind dies Böden mit einem geringen Sandanteil. Baue mit diesen Eigenschaften sind als suboptimal zu werten. Auch wenn einige davon im Laufe der Jahre Ausmaße erreichen, die an die Größe von optimalen Sandbauen heranrei- chen, so sind hier große Baue ım Vergleich zur Häufigkeit großer Baue in Sandböden selten. Danksagung Diese Untersuchung wurde z.T. durch Mittel des Bayerischen Staatsministeriums für Landesentwick- lung und Umweltfragen finanziert. Herrn Dr. H. H. BECHER vom Lehrstuhl für Bodenkunde der TU München danke ich für die Analyse der Bodenproben und Herrn E. LANGENSCHEIDT von der Nationalparkverwaltung Berchtesgaden für geologische Auskünfte im Gelände. Den Herren Dr. W. ARNOLD und Dr. F. TRILLMICH vom Max-Planck-Institut für Verhaltensphysiologie Seewiesen bin ich für kritische Durchsicht des Manuskripts dankbar. Zusammenfassung In zwei Untersuchungsgebieten Südostbayerns wurden 148 Dachsbaue hinsichtlich ihrer Eignung zur Bauanlage untersucht. Die Baugröße wurde getrennt nach Erd- und Felsbauen ermittelt. Anhand des vor den Bauen liegenden Erdauswurfmaterials wurde die Grabbarkeit des Bodens in vier Grabbarkeitsklassen ein- geteilt, worin sich die beiden Untersuchungsgebiete signifikant unterscheiden (x? =69,507, p < 0,001). Von 30 Erdbauen wurden Bodenanalysen durchgeführt, wobei u.a. die Körnung, die Bodenart und die Bonitierung des Bodens analysiert wurden. Nach dieser Bodenanalyse, mit deren Hilfe auch die anderen restlichen Böden zugeordnet werden konnten, wurden diese Böden in vier „Qualitätsstu- fen“ eingeteilt und diese mit der Baugröße in Beziehung gebracht. Daraus ergibt sich klar, daß die Dachse Sandböden vor allen anderen Böden bevorzugen (X? = 36,86, p <0,001) und hier auch die größsten Baue anlegen (H-Test, p < 0,001). Zum Graben eines Dachsbaues eignet sıch ein Boden von ungefähr der folgenden Körnung am besten: bis 10 % Tonanteile, 15 % Schluff und ca. 75 % Sand. Die Typen der Felsbaue und ihre Funktion werden bewertet und die ermittelten Befunde diskutiert. Die Daten sind mit der eingangs aufgestellten Hypothese, daß die unterschiedlichen Baugrößen beider Untersuchungsgebiete auf „Qualitätsunterschiede“ der Böden zurückzuführen sind, verträglich. Literatur Bayerisches Geologisches Landesamt (Hrsg.) (1981): Geologische Karte von Bayern 1:500000. München. pp. 5-168. Bundesanstalt für Geowissenschaften und Rohstoffe und den geol. Landesämtern in der Bundesrepu- blik Deutschland (Hrsg.) (1982): Bodenkundliche Kartieranleitung. 3. Aufl. Hannover. Bock, W. F. (1986): Kriterien zur Größenbeurteilung der Baue des Dachses Meles meles L. Säugetierkundl. Mitt. 33, 227-234. CLEMENTS, E. D. (1974): National badger survey in Sussex. Sussex Trust for Nat. Cons., Mamm. Report. for 1970/71. Dunwerr, M.R.; KiLLıngLey, A. (1969): The distribution of badger sets in relation to the geology o the Chilterns. J. Zool. 158, 204-208. GOETHE, F. (1955): Die Säugetiere des Teutoburgerwaldes und des Lipperlandes. Abh. Landesmus. Naturk. Münster 17, 5-19. Harrıs, $. (1984): Ecology of urban badgers Meles meles. Distribution in Britain and habitat selection, persecution, food and damage in the city of Bristol. Biological Conservatum 28, 349-375. Die Bedeutung des Untergrundes für die Größe von Banen des Dachses 99% HEPTNER, V. G.; Naumov, N. P. (Hrsg.) (1974): Die Säugetiere der Sowjetunion. Bd. II: Seekühe und Raubtiere. Jena: VEB G. Fischer. Kruuk, H. (1978): Spatial organization and territorial behaviour of the European badger Meles meles. J. Zool. (Lond.) 184, 1-19. LIKHACHEV, G. N. (1956): Some ecological traits of the badger of the Tula abatis Broadleaf Forests. Studies on mammals in Government Preserves. Ed. by YuRGENson. Moscov: Ministry of Agriculture of the USSR. LuGerT, J. (1985): Untersuchungen zur Bestandsentwicklung und Situation des Dachses (Meles meles Linne, 1758) in Schleswig-Holstein. Dipl.-Arbeit Univ. Kiel, Forschungsstelle Wildbiologie. NEAL, E. G.; HARRISON, R. J. (1958): Reproduction in the European Badgers (Meles meles). Trans. 100\3Soe. Lond. 29, 67-131. Near, E. (1972): The national badger survey. Mammal Rev. 2, 55-64. NEAL, E. G. (1977): Badgers. Poole: Dorset: Blandford Press. STUBBE, M. (1982): Schutz und Hege des Dachses. In: Buch der Hege. Hrsg. von H. STUBBE. Berlin: VEB Deutscher Landwirtschaftsverlag, pp. 294-317. WIJNGAARDEN, A. VAN; PEPPEL, ]J. VAN DE (1964): The badger Meles meles (L.), in the Netherlands. Lutra 6, 1-60. Anschrift des Verfassers: Dr. WALTER FRIEDRICH Bock, Institut für Haustierkunde, Universität Kiel, Olshausenstr. 40-60, D-2300 Kiel Z. Säugetierkunde 53 (1988) 358-364 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Scent marking behaviour in captive Water mongooses (Atılax paludinosus) By CaroLyNn M. BAKER Department of Biology, Natal University, Durban Receipt of Ms. 1. 7. 1987 Abstract Scent marking in captive Atzlax paludinosus ıs described. The four methods recorded include defaecation, urination, anal marking and cheek rubbing. Anal marking was most frequently used and in females the frequency was significantly higher during the non-breeding season. Sexual dımorphism with respect to anal marking was exhibited, females marking more frequently than males. Cheek rubbing frequency was higher in the presence of conspecifics and females showed a significant increase in cheek marking during the breeding season. Defaecation rates varıed with food intake, while urination frequencies were significantly higher when mongooses shared an enclosure with a con- specific. The relevance of the different rates and patterns of marking are discussed in relation to the ecology of solitary Atzlax, and comparisons with sociable Helogale are made. Introduction Pocock (1916) was the fırst to describe the external structure of the anal sac ın Atzlax. More recently interest in mongoose scent glands and their secretions has increased. GoRMAN (1976, 1980) and GoRMman et al. (1974) have analysed the anal secretions of Herpestes auropunctatus, and HEFETZ et al. (1984) have shown that Herpestes ichneumon exhibits sex specificity with regard to its anal gland secretions. In addition marking behaviour has been studied in Felogale undulata rufnla (Rasa 1973) and Suricata suricatta (MorAN and SORENSEN 1980). HEDIGER (1949), FIEDLER (1957), MicHAELIs (1972) and Ewer (1973) have described some of the marking behaviour in a variety of herpestines, including Atılax. The current study describes ın detail marking behaviour in captıve water mongooses. An attempt is made to interpret the meanıng of the various marking behaviour patterns and to compare these with patterns ın other herpestines. Material and methods Eight mongooses (three males and five females) were held captive in outdoor enclosures. Details regarding their housing are presented in BAKER and MEESTER (1986). Observations were made from an observation hide one metre distant from the cages. Marking sequences were either recorded on detailed checksheets or filmed using a National VHS portable video system. Frequency of marking events was analysed for each hour of observation. Sequence analysis was carried out with the aid of transition matrices and flow diagrams were constructed (CHATFIELD and LEMON 1970; LEHNER 1979). A behaviour sequence was delimited by the initial approach to and withdrawal from the object to be marked. Results Four methods of depositing scent products were recorded, including defaecation, urina- tion, anal dragging and cheek rubbing. Figure 1 shows the frequency of the different marking patterns. Allomarking was not observed and all marks were deposited on inanımate objects. However, hand-raised mongooses did mark their ‘human parent’. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5306-0358 $ 02.50/0 Scent marking behaviour in captive Water mongooses (Atilax paludinosus) 359 Defaecation In 15 different 24-hour periods defaecation occurred twice per individual animal. Mon- gooses tended to defaecate soon after becoming active in the early evening (between 17h30 and 19h30) and then again in the early hours of the morning (between 05h00 and 06h30) before retiring for the day. Scats were always deposited on specific scat sites or middens. In six out of the eight mongoose enclosures the middens were located at the furthest point from the nestbox, while in the remaining two enclosures the middens were positioned close to the nest site. All defaecation sites were located at ground level and scats were never deposited over tree trunks or on rocks. When two or more animals were housed together, a common midden was used. Once the particular site for defaecation was chosen there was never any change in its location. Defaecation postures usually involved a simple squat. The tail position was characteris- tic, extending straight backwards with a slight upward curve along its length. Mongooses often circled a few times before settling to defaecate. Urination Urination frequencies varıed from zero to 5 times per hour (X=0.55, n=158). Urination frequencies were significantly higher when mongooses were housed alone than when they shared an enclosure with a conspecific (d = 2.47, 0.02 > p >0.01). Urination sites varıed and, unlike scat sites, different parts of the enclosures were used, although some sites were preferentially marked such as small rocks, the midden and logs. Urination was usually performed in a squatting position with the base of the tail positioned slightly higher than during defaecation. Male mongooses regularly lifted a hınd leg when urinating, and on occasıon females were also noted to use the leg lift. When using the leg lift position the mongooses were apparently not aiming to increase the height of the urıne mark, as they did not always lift their legs next to a vertical object. ‚Anal dragging Frequency of anal marking varıed from zero through 21 times per hour (X= 1.96, n=143). The presence of known conspecifics had no significant effect on the rate of anal dragging (d=0.09, p >0.1). During the breeding season it was found that while there was no significant difference in anal marking frequency in males (d=1.32, p >0.1) there was a significant difference in females (d = 4.02, 0.0001 > p >0.00001) with more marking exhibited during non-breeding. Anal scent was deposited on small rocks, nest boxes, tree stumps and tunnels within the enclosures. In time it was possible to locate anal marking sites visually as a result of discolouration of the marked object, as more and more layers of scent were deposited. The anal drag was made by depressing the opened anal sac onto an object, as was described ın EwER (1973). The use of the handstand position as recorded by HEDIGER (1949) and EwEr (1973) was observed only three times during this study. On all occasions the mongoose concerned marked the upright walls of the nestbox after introduction to a new enclosure. The substance that was deposited had an oıly consistency and was black when initially deposited. As time elapsed and the amount of secretion built up as a result of repeated marking activity, the deposit became creamy. On occasion the mongooses forcibly ejected anal fluid from the anal sacs, usually when startled or ın stressful situations. This fluid was black, strong-smelling and extremely volatile. Inıtially the odour was overpowering, but over a period of 24 h ıt dissipated and was noticeable only if the deposit was closely investigated. After three days the smell was no longer distinguishable by the observer. 360 Carolyn M. Baker Cheek marks ] Urination A Anal marks NNDE ONE 2 3 4 5 6 7 8 SEE OBEN PrEISSnZEES NUMBER OF EVENTS PER HOUR Fig. 1. Frequency of scent marking events in Atılax Cheek rubbing Cheek rubbing frequency varıed from zero through three times per hour (x = 0.17, n = 142). Frequency of cheek rubbing was significantly higher in the presence of known conspecifics (d = 2.89, 0.01 > p >0.001). During the breeding season females marked significantly more than during the non-breeding season (d = 3.31, 0.001 > p >0.0001), while ın males no significant difference was found (d = 0.9, p >0.1). The most commonly rubbed objects included the entrance to the nestbox, tree stumps and tunnels. Cheek in. involved slowly wiping the sides of the head, from the corners of the mouth to the base of the ear pinnae, against an object. The secretion was never visible and the marked object bore no humanly- neridessik smell. However, the cheeks of the mongooses did have a pleasant honey-like odour. Figure 2 shows the sequences of behaviour patterns followed during glandular marking. Twelve transitions are shown, Scent marking behaviour in captive Water mongooses (Atılax paludinosus) 361 indicating that the most common ac- tions involved ınitially smelling the object to be marked, followed by a0 either cheek or anal marking. The observed transition matrix was sig- nificantly different from a random , VE ; | = distabution, y2= 171.23, d.f.=16, BEGIN SNIFF ars END p<0.001. Mean duration of se- ur we F quences was 11.61 sec, n = 40, SD = I ORZEDE \ 0,01 Discussion 5 The most frequent Scan marking Fig. 2. Flow diagram showing the sequence of be- pattern used by Atılax ıs anal marking, havioural events during scent marking by Atilax. followed by urination, cheek marking * Transitions above expected values and then defaecation. Clearly the amount of faecal and urinary material produced by a mongoose is dependent on its dietary intake. This may affect the frequency of defaecation and urination. In a captive environment, anımals receive a constant and usually abundant supply of nutrients, so that the results reported here in respect of these two functions may bear little resemblance to the situation in naturally occurring Atilax. However, urination appears to serve not only as a means of excretion but also as a marking method. This is demonstrated by the high frequency of this activity, and the fact that only a few drops of fluid need be produced at a time. The use of middens by carnıvores ıs widespread (MAcponALD 1980). While middens in this study were never located on rocks, middens in the field are regularly found on rocky outcrops and bare expanses of rock as well as on sandy river banks. The lack of rock use by captive anımals probably resulted from the absence of suitably large rocky surfaces. In Atilax midden use may be the most efficient method of information dissemination, regarding the inhabitants of a particular area. If, as EwWER (1973) suggests, the faeces are coated with anal gland secretions as they pass out of the body, the scats become the calling card of that particular individual. GoRMAN (1976) has shown that individual Herpestes auropunctatus can recognize the anal gland secretions of particular conspecifics, and HEFETZ et al. (1984) have shown that Herpestes ichneumon males produce an anal scent component that is specific to their sex. Rasa (1973) has also shown that dwarf mongoose scats bear the identity of the individual but that the identifying scent of faeces does not persist for long periods. If we are to assume that the same characteristics of faecal material apply to water mongooses then the importance of regular use of the same middens becomes apparent; by continuously using the same midden the scent of an anımal is renewed regularly enough to inform other conspecifics of its presence. The infrequent use recorded during this study of the handstand position in anal scent deposition may result from an insufficient number of appropriate upright marking posts. Further, as handstand marking was performed only when a mongoose was introduced to a new enclosure, it may be concluded that handstand marking is used only in those circumstances when the mongooses are highly motivated to mark. Although cheek glands have not been anatomically described for this species, their presence is indicated on behavioural grounds both here and elsewhere (ZANNIER 1965; Rasa 1973). The marking sequences show a clear relationship between cheek marking and sniffing. As cheek marking was usually (50.7 % of the time) preceded by sniffing, it seems that the mongooses initially ascertained the status of the previous mark before either renewing their 362 Carolyn M. Baker own or covering that of a conspecific. Most cheek marking behaviour (96 %) occurred when mongooses were housed with a conspecific of the opposite sex during the breeding season. During these times levels of agonistic behaviour were high and 71.4 % of the cheek marking episodes were carried out by females. If we assume that cheek marking carries a message of excitement/threat as in dwarf mongooses (Rasa 1973), then the incidence of cheek marking during premating encounters is understandable, as during this period the female wards off and avoids the male. Only when she is ready to copulate does she cease cheek marking and entirely submit to her mate’s demands. In Helogale (Rasa 1973) cheek marking is preceded by sniffing 71.2 % of the time, while anal marking is shown to mainly follow cheek marking. In Atilax, on the other hand, anal marking was preceded by sniffing 37.3 % of the time and never followed cheek marking. In addition Atzlax preceded most marking sequences by sniffing (88 % of the time), indicating the importance of scent marking in their communication. Rasa (1973) has shown that cheek marks in Felogale have a brief effective period when compared with anal marks. In view of the sociability of dwarf mongooses, based on kinship and the concomitant need to maintain cohesion in the group, cheek marks may play a vital role in dissuading potential conspecific intruders from attempting to join the group, and it is therefore essential to precede most marking bouts with cheek marks. Rasa (pers. comm.) indicates that cheek marking plays no role in establishing and maintaining rank within the famıly group. For solitary Atılax a sıtuation in which conspecifics would want to become part of a group does not arise, and hence cheek marking takes on a less significant role. However anal marking reaches far more significance when the need to advertise their identity to potential mates is considered. This is supported by the higher frequencies of anal marking when compared with cheek marking. Interestingly, licking follows only anal marking in Helogale (Rasa 1973) while ın Atılax licking followed only cheek marking. Rasa (1973) suggests that the mongooses occasion- ally lick anal secretions because of the general interest engendered by these secretions as well as their lack of threat connotation. In Atılax the cheek marks that were licked belonged to the owner and licking always immediately followed the cheek mark. In these cases licking may serve to reassure the mongoose of the intended threatening message. When water mongooses were presented with a strong-smelling compound such as a deodorant or insecticide they frantically cheek-marked the object, rather than anal-marked it. In these circumstances (which are clearly unlikely to occur within the natural environ- ment) cheek marking may have no threat connotation, and is more likely to be released by high excitation levels. In general there was an even distribution of anal marking regardless of the presence of conspecifics, although 58 % of marking sequences occurred when conspecifics were present. Considering the identifying function of anal marking it is to be expected that ın the natural environment it should occur more regularly as ıt would inform conspecifics of their presence. The major sources of contact with conspecific water mongooses would be along feeding routes and at middens, and it is for this reason that anal scent is important as discussed earlier. To be able to establish the most recent visit to a midden by a conspecific would convey information regarding potential mates in the area. This is supported by the significant difference in marking frequency in females during the breeding season. In sociable mongooses the use of middens is likely to play a less crucial role with regard to breeding, as the members of the group are ın constant contact anyway. Stoats, like water mongooses, are solitary animals. ErLINGE et al. (1982) has shown that dominance in these mustelids is conveyed by increased marking levels and that subordinate animals attempt to avoid or escape from the dominant partner. Water mongooses exhibited similar behaviour patterns when housed in pairs during the breeding season, in that Scent marking behaviour in captive Water mongooses (Atılax paludinosus) 363 initially the male was dominant and the female exhibited avoidance and escape reactions. As the female approached readiness to copulate the dominance relationship reverses and the female cheek-marked more frequently, initiated allogrooming and ceased avoidance behaviour. Rasa (1973) has shown that dominance status is not related to marking frequency ın Helogale, but rather that age and sex: of mongooses reflect marking fre- quency. In this study Atılax exhibited sexual dımorphism with regard to anal marking, with females marking significantly more frequently (d = 6.97, p <0.00001). In Helogale the reverse situation occurs wıth males marking more frequently with the anal glands than females (Rasa 1973), and ıt is suggested that this may sımply reflect sexual behaviour differences. However, cheek marking exhibited no such dimorphism in Atılax (d = 1.21, p > 0.1). ERLINGE et al. (1982) shows that stoats show no consistent difference in marking patterns ın males and females, and goes further to say that this is expected in view of their spacing pattern. Stoats are solitary mustelids in which resident males and females exclude members of their own sex from their territories. Because the spacing pattern of Atzlax in the natural environment has not been reported, no direct relationship between marking and territoriality can be made. In addition ıt has been noted that to assess direct relationships between marking and territorial behaviour is of limited value (BARRETTE and MessıEr 1980) as marking may simply be a means of expressing social status and reproductive condition (BEKOFF and DıamonD 1976) or of familiarızing an animal with its environment (EwER 1968). As males showed no significant difference in marking patterns throughout the year, it is suggested that they were transıtory with regard to territories, while females maintained loose territories and advertised their presence and condition by varying marking fre- quency. The fact that females anal-marked less frequently during the breeding season and cheek-marked more frequently at this time, indicates that their excitement levels increased. By virtue of the increased intensity of cheek marking ın conjunction with anal marking, males could readily perceive the reproductive condition of females. KOEHLER et al. (1980) suggested that in wolverines marking may serve as a mechanısm of avoidance, and that this may be a general feature of many solitary anımals. When Atılax pairs were housed together for extensive periods avoidance by both mongooses was initially observable, but after a few weeks the anımals regularly inhabited the same nestbox. When active, however the tendency to avoid one another was evident until the onset of the breeding season. Scent marking in animals is clearly an important mode of communication that is influenced not only by motivational and circumstantial cues but also by physiological ones. In Atilax the use of scent is essential, especially in view of its solitary mode ot life. The results indicate that the function of scent varies, serving to promote avoidance when evidence of oestrous is absent and facilitating contact during the breeding season. Acknowledgements Thanks are extended to: Prof. J. MEESTER for supervising this project and critically reading the manuscript; Prof. G. MAcLEAN for translating the summary; Prof. A. Rasa for commenting on the manuscript; the Council for Scientific and Industrial Research and University of Natal Research Fund for financial assistance; ELEANOR RICHARDSON and GRAHAME Hayes for support and also for carıng for the anımals in my absence. 364 Carolyn M. Baker Zusammenfassung Duftmarkierung bei gefangengehaltenen Sumpfichneumons (Atilax paludinosus) Unterschiedliche Methoden der Duftmarkierung von Atzlax paludinosus in Gefangenschaft werden beschrieben. Die vier beobachteten Markierungsweisen sind Koten, Harnen, anale Markierung und Reiben der Wangen. Am häufigsten fand die anale Markierung statt, die bei Weibchen außerhalb der Fortpflanzungszeit noch häufiger war. Ferner ergab sich bei analer Markierung ein Geschlechtsunter- schied, denn Weibchen markierten öfter als Männchen. Die Häufigkeit der Markierung durch Wangenreiben nahm zu, wenn sich Tiere derselben Art zusammenfanden; während der Fortpflan- zungszeit markierten Weibchen signifikant mehr. Die Häufigkeit der Kotabgabe stand mit der Nahrungsaufnahme in enger Verbindung; Harnen nahm signifikant zu, wenn Mangusten sich mit Artgenossen zusammenfanden. Die Bedeutung dieser verschiedenen Markierungsarten und -muster wird mit Bezug auf die Okologie der einzellebenden Atzlax diskutiert, und mit dem Verhalten der geselligen Helogale verglichen. Literature BAKER, C. M.; MEESTER, J. (1986): Postnatal physical development of the water mongoose (Atilax paludinosus). Z. Säugetierkunde 51, 236-243. BARRETTE, C.; MEsSIER, F. (1980): Scent-marking in free-ranging coyotes, Canis latrans. Anim. Behav. 28, 814-819. BEKOFF, M.; DIAMOND, J. (1976): Precopulatory and copulatory behaviour in coyotes. J. Mammalogy 37, 312-375. CHATFIELD, C.; LEMON, R. E. (1970): Analysing sequences of behavioural events. J. theor. Biol. 29, 427445. ERLINGE, $.; SANDELL, M.; BRINckK, C. (1982): Scent-marking and its territorial significance in stoats, Mustela erminea. Anım. Behav. 30, 811-818. EwER, R. F. (1968): Ethology of mammals. London: Paul Elek (Scientific Books) Ltd. Ewer, R. F. (1973): The carnıvores. London: Weidenfeld and Nicolson. FIEDLER, W. (1957): Beobachtungen zum Markierungsverhalten einiger Säugetiere. Z. Säugetierkunde 22, 57-76. GORMAN, M. L. (1976): A mechanism for individual recognition by odour in Herpestes auropunctatus (Carnıvora: Viverridae). Anım. Behav. 24, 141-146. GORMAN, M.L. (1980): Sweaty mongooses and other smelly carnıvores. Symp. Zool. Soc. Lond. 45, 87-103. GoRrMman, M. L.; Nepweıı, D. B.; SMITH, R. M. (1974): An analysıs of the anal scent pockets of Herpestes auropunctatus (Carnıvora: Viverridae). J. Zool. Lond. 172, 389-399. HEDIGER, H. (1949): Säugetierterritorien und ihre Markierung. Bijdragen tot de dierkunde 28, 172-182. Heretz, A.; BEn-Yaacov, R.; Yom-Tov, Y. (1984): Sex specificity ın the anal gland secretion of the Egyptian mongoose (Herpestes ichneumon). J. Zool., Lond., 203, 205-209. KOEHLER, G. M.; HORNOcKER, M. G.; HasH, H. $. (1980): Wolverine marking behaviour. Canadian Field Naturalist 94, 339-341. LEHNER, P. N. (1979): Handbook of ethological methods. New York: Garland STPM Press. MaAcponALD, D. W. (1980): Patterns of scent-marking with urine and faeces amongst carnıvore communities. Symp. Zool. Soc. Lond. 45, 107-139. MicHaeuis, B. (1972): Die Schleichkatzen (Viverriden) Afrikas. Säugetierkdl. Mitt. 20, 1-110. Moran, G.; SORENSEN, L. (1986): Scent-marking behaviour in a captıve group of meerkats (Suricata suricatta). J. Mammalogy 67, 120-132. Pocock, R. I. (1916): On the external characters of the mongooses (Mungotidae). Proc. Zool. Soc. Lond. 1916, 349-374. Rasa, ©. A. E. (1973): Marking behaviour and its social significance in the African dwarf mongoose, Helogale undulata rufula. Z. Tierpsychol. 32, 293-318. ZANNIER, F. (1965): Verhaltensuntersuchungen an der Zwergmanguste, Helogale undulata rufula, ım Zoologischen Garten Frankfurt am Main. Z. Tierpsychol. 22, 672-695. Author’s address: CaroLyn M. Baker, Zoology Department, University of Durban-Westville, PB X 54001, Westville, R.S.A. Z. Säugetierkunde 53 (1988) 365-379 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Zur Anatomie des Gehörorganes der Seekuh (Trichechus manatusL.), (Mammalıa: Sirenia) VonM. S. FISCHER Eingang des Ms. 5. 8. 1987 Abstract Contributions to the anatomy of the hearing organ of the seacow Trichechus manatus L. (Mammalia: Sirenia) The anatomy of the soft parts of the hearing organ of Trichechus manatus L. is described. Two specimens have been dissected and in both cases the external auditory meatus ends in a blind pouch, which has no contact to the tympanic membrane. The latter is pushed outwards by the Manubrium mallei. The tympanıc membrane consists of unusual dense connective tissue. A major peculiarity of the middle ear of Trichechus is the tympanic sac. Ventrally the tympanic cavity is bordered by a membraneous sac. It encloses the space between the basioccipital, basisphenoid, the pterygoid process and the tympanic ring. The tympanic sac is part of the tympanic cavity and covered on the inner side with a respiratory epithelium. The Eustachian tube is highly divergent from the condition found in other mammals. A comparison of the tympanic sac with the Eustachian sac of Procaviidae, Equidae and Tapiridae reveals no homology between these two structures. Cautious reflexions on the function of the tympanic sacs suggest, that they could be a kind of air-cushion which isolate the ear against the sirenian’s own vocalızation. The problem of directional hearing under water and sound conduction is briefly discussed and applied to the morphology of the manatee. Einleitung Unsere derzeitigen Kenntnisse über das Gehörorgan der Sırenia beschränken sich fast ausnahmslos auf die Anatomie der Hartteilstrukturen (HyrtL 1845; CraupDıus 1867; Doran 1876; van KAMPEN 1905; MATTHES 1912; VAN DER KLauuw 1931; WERNER 1960; ROBINEAU 1965, 1969; FLEISCHER 1971, 1973, 1976, 1978) und deren Ontogenese (DILG 1909; FREUND 1908; MATTHES 1921; HIRSCHFELDER 1936). Die umfassendste Darstellung der Skelettstrukturen findet sıch in den Arbeiten von ROBINEAU. Das Problem des Hörens bei aquatischen Säugetieren hat in den letzten Jahrzehnten großes Interesse erregt, und es sind wichtige Arbeiten hierzu veröffentlicht worden, die zumeist die Anatomie und Hörphysiologie der Wale betreffen (z.B. REYSENBACH DE Haan 1957; REPENNING 1972; Purves und Pırrerr 1983). Der Weichteilanatomie des Ohres der Seekühe ist hingegen keine Aufmerksamkeit zuteil geworden. Leider sind ın Arbeiten von FLEISCHER (1971-1978) die anatomischen Verhältnisse nur schematisch wiedergegeben. Da kürzlich neue Befunde über das Hörvermögen der Seekühe bekannt geworden sind (BuLLock et al. 1982), die die Möglichkeit des Ultraschallhörens auch bei dieser aquatischen Gruppe in Betracht ziehen, erschien eine anatomische Untersuchung des Gehörorganes der Seekühe umso wünschenswerter. Die Kenntnis der Anatomie der Weichteile dieser Region ist in einem weiteren Zusammenhang von Bedeutung. Dem Auftreten von Diverticula tubae Eustachii kommt möglicherweise ein besonderer Wert für die Systematik der Perissodactyla (Mesaxonia + Hyracoidea) zu (FISCHER 1986). Da Murıe (1874) einen „Eustachian sac“ bei Trichechus manatus beschrieb und bekanntlich von vielen Autoren eine enge Verwandtschaft zwi- schen Sirenia, Proboscidea und Hyracoıdea angenommen wird, muß somit auch geklärt werden, inwieweit eine Homologie zwischen den Luftsäcken der Perissodactyla und der Sirenia besteht. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5306-0365 $ 02.50/0 366 M. S. Fischer In der vorliegenden Arbeit will ich eine ausführliche Beschreibung der Anatomie des Mittelohres und insbesondere des Tympanalsackes von Trichechus manatus geben, und die Befunde mit den Verhältnissen der Cetacea und Perissodactyla vergleichen. Darüber hinaus soll versucht werden, Besonderheiten des Gehörorganes der Seekühe auf die Bedingungen des Hörens unter Wasser zu beziehen. Material und Methode Die Sektion wurde an einem Exemplar der Seekuh Trichechus manatus L. durchgeführt, die Prof. STARCK für die Sammlung der Senckenbergischen Anatomie 1975 vom Zoologischen Garten Duisburg erhalten hatte. Das Alter des Tieres ist unbekannt, da es sich um einen Wildfang aus Guyana handelte. Abmessungen: Gesamtlänge (Schnauzenspitze — Schwanzflossenspitze) = 158 cm, Länge Schnau- zenspitze — Analöffnung = 103 cm, Länge Schnauzenspitze — äußere Gehörgangsöffnung = 22,5 cm, Leibesumfang in Höhe des Nabels = 107 cm. Das Tier wurde über die Schwanzarterie mit 4%igem Formalin perfundiert und in 4 %igem Formalın aufbewahrt. Die makroskopische Präparation wurde durch histologische Untersuchungen des Tympanalsackes, des Trommeltelles, des Endabschnittes des äußeren Gehörganges und verschie- dener Abschnitte der Tuba Eustachii ergänzt. Der Erhaltungszustand der Gewebe war nicht immer befriedigend. An einem juvenilen Tier, welches die zoologische Schausammlung der Universität Tübingen kurz vor Abschluß der Arbeit aus Florida erhielt, wurde eine Kontrollsektion des äußeren Gehörganges durchgeführt. Ergebnisse Meatus acusticus externus Die winzige äußere Ohrötfnung (Abb. 1), durch die, wie schon Mur1eE (1874) bemerkte, gerade eine Sonde eingeführt werden kann, liegt 11 cm hinter dem Auge. Der Gehörgang (Meatus acusticus externus) durchbricht die an dieser Stelle 14 mm dicke Haut und nımmt Abb. 1. Die äußere Ohröffnung von Trichechus manatus. Die äußere Ohröffnung ist von der sie umgebenden Haut nur durch die hellere Färbung zu unterscheiden Zur Anatomie des Gehörorganes der Seekuh (Trichechus manatus L.) 367 einen caudoventralen Verlauf. Er erweitert sich zu einem Sack, der überraschenderweise blind endet (Abb. 2). Der Blindsack hat keinen Kontakt zum Trommelfell. Die Länge des äußeren Gehörganges ist 53 mm. Der Durchmesser des Blindsacklumens beträgt maximal 5mm und ıst damit etwa doppelt so groß wie das Lumen des Gehörganges. Aus dem medialen Teil der äußersten bindegewebigen Umhüllung des Meatus acusticus externus Abb. 2. Trichechus manatus. Die Zeichnung zeigt die Lage des äußeren Gehörganges (Mea). Er durchbricht zunächst die dicke Haut und endet in Höhe des Jochbogens (J) in einem Blindsack, der keine Verbindung zum Trommeltell besitzt. Die Ohröffnung (Porus acusticus) ist durch einen Pfeil markiert geht ein Band hervor, das den Unterrand des Blindsackes fixiert, sich flächenartig verbreitert und am ventralen Rand des Tympanıcum ansetzt. Dieser Befund konnte an dem erwähnten juvenilen Exemplar von Trichechus manatus der Zoologischen Schausammlung der Universität Tübingen überprüft werden. Auch bei diesem Exemplar endete der äußere Gehörgang in einem Blindsack. Dies steht im Widerspruch zu der Beobachtung von MurıeE (1874), der einzigen, die ıch in der Literatur gefunden habe. Er beschreibt den äußeren Gehörgang als „a narrow cord-like fibrous tube, 3 inches long, with an S-shaped bend, leads to the membrana tympanı“ (Murıe 1874, p. 188). Ob der äußere Gehörgang bei den Seekühen eine Funktion beim Hören hat, scheint sehr fraglich zu sein. An der Lumenseite ist der äußere Gehörgang mit einer dunklen, stark pigmentierten Schicht ausgekleidet. Die histologische Untersuchung zeigt, daß es sich um ein unverhorn- tes mehrschichtiges Plattenepithel handelt, und daf die Pigmentzellen in den unteren Epithelschichten liegen. Des weiteren sieht man im Querschnitt mehrere Lagen konzen- trisch angeordneter, abwechselnd longitudinal und circulär ziehender Bindegewebstfasern. Es finden sich keine Drüsen. Die mirkoskopische Untersuchung bestätigt auch, daß der Meatus acusticus externus ın der Tiefe allmählich sein Lumen verliert und blind endet. Membrana tympani Schon lange ist bekannt, daß das Trommeltell der Seekühe, verglichen mit anderen Säugetieren, eine einzigartige Form besitzt. Es wırd vom Manubrium malleı, das dabei die Form eines leicht nach rostral gekippten Ringes hat, zeltartig nach lateral vorgebuchtet 368 M. S. Fischer (Abb. 3, 4). Das Trommelfell ist in einem seichten Sulcus tympanicus befestigt. Dorsal des Manubrium mallei schlägt es nach innen um und inseriert am Innenrand des Malleus. Oberhalb des vorderen Schenkels des Tympanicum dehnt sich das Trommelfell rostrad aus. Auch dieser Teil des Trommelfelles wendet sich scharf nach innen um. Er ist an der medialen Seite des Caput malleı befestigt. Die Grenze zwischen dem gespannten und dem schlaffen Teil des Trommeltfelles verläuft vom höchsten Punkt des vorderen Schenkels schräg nach dorsal. Sie ist als weiße Linie in Abb. 3 eingezeichnet. Ob dieser schlaffere Teil des Trommeltelles der Pars flaccıda (Shrapnellsche Membran) entspricht, wie ROBINEAU (1969) annımmt, kann hier nicht entschieden werden. Abb. 3. Trichechus manatus. Lateralansicht der Tympanalregion. Das in dem hufeisenförmigen Anulus tympanicus aufgespannte, derbe Trommelfell wird vom Manubrium mallei (Mm) nach außen gedrückt (vgl. Abb. 4). Der kräftige N. facıalis (N. fac) verläßt das Cavum tympanı zwischen Anulus tympanicus und dem knorpeligen Teil des Stylohyale (Sty). Ventral des Trommeltellringes erkennt man den lateralen Teil des Tympanalsack Der Durchmesser des straff gespannten Trommelfelles beträgt ın caudorostraler Rich- tung 19 mm und dorsoventral 24 mm. FLEISCHER (1971) gibt den Durchmesser bei 2 Exemplaren von Dugong dugon mit 16 mm und 19 mm an. Nach FLEISCHER (1978) beträgt der Durchmesser des Trommelfelles der Sirenia durchschnittlich 20 mm. Der Dünnschnitt ergibt eine Dicke des Trommelfelles von 0,6 mm. Die histologische Untersuchung zeigt, daß das Trommelfell von einer ungewöhnlich derben, wohlgeschichteten Bindegewebsplatte gebildet wird. Das Bindegewebe zeigt geordnete längs- und querverlaufende Züge, die auf eine sehr starke Verflechtung schließen lassen. Unterhalb des mehrreihigen Epithels, das das Trommelfell zum Cavum tympanı hin bedeckt, finden sich wenige muköse und seröse Drüsen. Ein äußeres Epithel ist wegen der fehlenden Verbindung mit dem äußeren Gehörgang nicht mehr vorhanden. Dem Trommeltell schließt sich hier ein lockeres Bindegewebe mit elastischen Elementen an. Hyrrı (1845) und Doran (1876) beschreiben, daß das Trommeltell bei Dugong dugon nicht wie bei Tyichechus durch das Manubrium malleı lateral vorgestülpt, sondern plan seı. Zur Anatomie des Gehörorganes der Seekuh (Trichechus manatus L.) 369 Hydrodamalıs gigas summe dagegen mit den Verhältnissen bei Trichechus überein. Die Beobachtung an 9 Schädeln aus den drei Gattungen der Sırenıa mıt den Gehörknöchelchen in situ, die ich im British Museum (Natural History) untersuchte, legt die Vermutung nahe, dafs der Malleus durch die Mazeration seine distale Verbindung zum Anulus tympanicus verliert. Infolge des Verlustes dieser Befestigung könnte der Malleus durch die Spannung des Trommelfelles in die Paukenhöhle zurückgedrängt werden. So ist mög- licherweise der beoboachtete Unterschied zwischen Dugong und Trichechus auf ein Mazerationsartefakt zurückzuführen. Es sei auch noch bemerkt, daß das von FLEISCHER (1971) beobachtete Knochenplättchen am rostroventralen Rand des Anulus tympanicus bei den von mir untersuchten Exemplaren nicht auftrat. Tympanalsack Bei den Sirenia wird medial des Anulus tympanicus der Boden der Paukenhöhle von einem membranösen Sack gebildet. Da dieser Sack und die Paukenhöhle offen miteinander kommunizieren und von der selben Schleimhaut ausgekleidet werden, ist davon auszuge- hen, daß es sich um eine Vergrößerung des Cavum tympani handelt. In der Literatur finden sich nur zwei kurzgefaßte Bemerkungen von Murıe (1874) und ROBINEAU (1965) über den Tympanalsack. Sie beschränken sich darauf seine Lage zwi- schen Basısphenoid, Basioccipitale, Exoccipitale und Tympanicum zu beschreiben. Eine von FISCHER (1986) wiedergegebene briefliche Mitteilung von D. P. Donntng enthält die zusätzliche Information, daß der Tympanalsack in Höhe des Processus pterygoideus in die ziemlich große Tuba Eustachii übergehe. Die Tympanalsackmembran löst sich medial vom Anulus tympanıcus ab (Abb. 3, 4), der so die laterale Begrenzung des Tympanalsackes bildet. Caudal liegt der Tympanalsack dem knorpeligen Beginn des Zungenbeines an. Vom M. brachiocephalicus bedeckt, zieht er zu den Austrittsstellen der Hirnnerven IX-XII, auf deren Höhe sich die Membran scharf nach rostral wendet und eine Duplikatur bildet, die die A. carotis interna und den N. caroticus internus vom Eintritt in den Tympanalsackraum bis zum Übergang in den Schädel umgibt. In Abb. 4 und Abb. 6 ist die Richtung der Duplikatur (Dupl.) wiederge- geben. Im Anschluß an diese Faltenbildung verläuft die Grenze des Tympanalsackes ent- lang des lateralen Randes der Schädelbasisknochen Basioccipitale und Basısphenoid und um die Insertion des M. longus capitis und des M. rectus capitis ventralis herum. Unmittelbar rostral der Ansatzstelle dieser beiden Kopfbeuger berühren sich die Tympa- nalsäcke der rechten und linken Seite in der Medianebene. Rostroventral stoßen sie an den Bindgewebsblock, der den pharyngealen Abschnitt der Tuba Eustachii umgibt. Rostrola- teral liegt der M. tensor velı palatını der Tympanalsackwand an. Die dorsale Wand des Tympanalraumes ist zugleich die ventrale Abdeckung der Fissura sphenopetrose (Foramen lacerum medium). Die Tympanalsackmembran ıst zwischen der Schädelbasis und dem Petrosum ausgespannt. Die in Abb. 4 eingezeichnete Schnittebene (S) zeigt, daß der Mediosagittalschnitt (Abb. 5) den Tympanalsack geringfügig parasagıttal getroffen hat, so daß man ihn im Anschnitt sieht. In Höhe ihrer Berührungstlächen befinden sich die Tympanalsäcke ventral des Tuberculum musculare an der Synchondrosis sphenooccipita- lis. Der Ventralseite des Tympanalsackes liegt ein kräftiger Fettkörper an, der sich auch zwischen seine Wand und den M. stylopharyngeus einschiebt. Er wurde in Abb. 5 nicht eingezeichnet, um die Eintrittstelle der A. carotis interna und des N. caroticus internus ın die ventrale Tympanalsackwand darzustellen. Nach Öffnung des Tympanalsackes auf der Ventralseite (Abb. 6) sieht man, daß der Tympanalsackraum durch zwei sich gegenüberstehende sichelförmige Falten, die eine elliptische Öffnung freilassen, in zwei Kompartimente unterteilt wird. Die beiden Falten (v.F. + h. F.) liegen in der Verlaufsrichtung der Tuba Eustachii und können am Schädel als caudale Fortsetzung der Kante gedacht werden, die selbst als auslaufende Lamina medıalıs 370 M. S. Fischer Mm N.fac Sq D.art N.mas ct N. lin M.stm Sty A.cin N.cin Dupl M.bra IX,X, X1,XIl M.rcv Abb. 4. Trichechus manatus. Die Präparation der Ventralseite der Region zwischen Tympanicum und Schädelbasis zeigt die Größe der Tympanalsäcke. Sie berühren sich in der Medianebene. Ausgehend von den Hirnnerven IX—XII ist die Duplikaturfalte eingezeichnet, in welche die A. carotis interna eingehüllt ist. Die Duplikatur unterteilt den von den Tympanalsackmembran umschlossenen Raum. Die gestrichelte Linie markiert die leicht parasagittale Schnittführung der in Abb. 5 gegebenen Ansicht proc. pterygoidei aufzufassen ist. Die hintere Falte ergibt sich aus der beschriebenen Duplikatur der caudalen Tympanalsackmembran. Sie endet rostral 1-2 cm vor der Ein- trittsstelle der A. carotis interna (Abb. 6). Zwischen der hinteren Falte und dem rostralen Rande des Anulus tympanicus bleibt ein spalttörmiger, hoher Durchtritt zum lateralen Tympanalsackraum und demjenigen zwischen Tympanıcum und Petrosum. Ob hier ein Verschluß in der Art eines Klappenventiles des medialen gegen den lateralen Raum vorgenommen werden kann, ist am toten Tier nicht zu klären. Caudal der A. stapedialis finden sich Ausbuchtungen in dem Teil der Tympanalsack- membran, der vom ventrocaudalen Rand des Petrosum ausgeht und die laterale Wand der Duplikatur bildet, die die A. carotis interna und A. stapedialis umhüllt (Abb. 6). Es treten 3 größere und 5 kleine Ausbuchtungen (Cellulae tympanicae?) auf. Im Anschluß an die am weitesten dorsocaudal gelegene Ausbuchtung bildet das Cavum tympani eine tiefe Aus- höhlung zwischen Petrosum und Exoccipitale. Auch rostral findet sich eine geräumige membranöse Ausbuchtung zwischen Petrosum und der Ventromedialseite des Anulus tympanicus. D. P. Domntng beobachtete bei zwei Exemplaren von Trichechus inungunis keine Ausbuchtungen im caudalen Abschnitt (Domninc briefl. Mittl. v. 19. 6. 1987). Trotz der für histologische Untersuchungen mangelhaften Erhaltung des Präparates erkennt man auf dem mikroskopischen Präparat noch vereinzelt Kinocilien an dem mehrreihigen, mit wenigen Becherzellen besetzten Epithel, das den Tympanalsack ausklei- det. Dieses respiratorische Epithel wird von einer drüsen-, gefäß- und nervenreichen Lamina propria unterlagert. Die Zahl der mukösen und serösen Drüsen ist im Bereich des Anulus tympanicus deutlich geringer als in der Schleimhaut der Tympanalsackmembran des beschriebenen Nebenraumes. Die Membran besteht im weiteren aus einem unregelmä- Rigen, straffen Bindegewebe mit wechselnder Schichtung. Die Dicke der gesamten Mem- bran beträgt 0,4-0,5 mm. Zur Anatomie des Gehörorganes der Seekuh (Trichechus manatus L.) 371 Opta Hy M.lca N.cin Rpha L.apd A.cin IX Xu M.rcv BD RS > © VER (FE X > BUS daR NH d> N.las /A.cco /T.sym M.cap Oes „ Al M.cop Abb. 5. Trichechus manatus. Der Mediosagittalschnitt zeigt die Lage des Tympanalsackes (Tys) zwischen wirbelseitiger Rachenwand, Schädelbasis und M. stylopharyngeus (M. stp). Der M. longus capitis (M. Ica) ist am Ansatz abgeschnitten. Die Tuba Eustachii (Pfeil) öffnet sich nach nur kurzem Verlauf in den Tympanalsack, lateral einer vorderen Falte (v. F.). Deutlich ist die Aufspaltung der A. carotıs communis (A. cco) zu sehen und der Eintritt der A. carotis interna (A. cin) in die ventrale Wand des Tympanalsackes. Man beachte die ungewöhnliche Form der Epiglottis Cavum tympani Da der Tympanalsack und die knöchernen Anteile des Cavum tympanı von derselben Schleimhaut ausgekleidet werden und der vom Tympanalsack umgebene Raum eindeutig zur Paukenhöhle gehört, entsteht eine Gliederung des Cavum tympanı in mehrere Kompartimente. In der Reihenfolge von medial nach lateral sind dies zunächst der medial der vorderen und hinteren Falte liegende Teil des Tympanalsackraumes. Daran schließt sich der durch die Faltenbildungen charakterisierte laterale Tympanalsackraum an, der sich rostral in die Tuba Eustachii fortsetzt. Auf das Problem der Abgrenzung der Tuba Eustachii vom Mittelohrraum wird weiter unten eingegangen. Die beiden Tympanalsack- räume könnten in Anlehnung an van KAampen (1905) hypotympanale Nebenhöhlen genannt werden. Der laterale Tympanalsackraum geht in den Raum über, der zwischen Anulus tympanicus und Trommelfell, Tegmen tympanı und Petrosum liegt und den man nach van KAmPEn als Paukenhöhle im engeren Sinne bezeichnet. Zu dieser gehört auch der sich unter dem kuppelförmigen Tegmen tympanı ausbreitende große Recessus epitym- panicus. Tuba Eustachii Über die Tuba Eustachii der Sirenia finden sich nur zwei kurze Angaben in der Literatur. „Ihe large Fustachian tube communicates with the auditory chamber just in front of the 572 M. S. Fischer ne A.sta A.cin ti TE xı,xıı Dupl > Abb. 6. Trichechus manatus. Ventrolateralansicht der Mittelohrregion. Das Trommeltell und ein Teil des Anulus tympanicus wurden entfernt. In die ventrale Tympanalsackmembran wurde ein großes Fenster geschnitten. Die innere Schleimhaut des Tympanalsackes geht ohne makroskopisch oder histologisch erkennbare Veränderung aus der das Cavum tympanı auskleidenden Schleimhaut hervor. An den Gehörknöcheln und am Tegmen tympani (Tt) wurde die Schleimhaut nicht eingezeichnet. Die erwähnte Duplikatur der Tympanalsackmembran bildet eine hintere Falte (h. F.), die zum Petrosum (P) zieht und in z. T. tiefe membranöse Kammern unterteilt ist. Durch eine elliptische Öffnung stehen die beiden von der Tympanalsackmembran umschlossenen, hypotympanalen Räume in Verbindung (Pfeil). Man beachte auch die dünne A. stapedialis (A. sta) und die ungewöhnliche Form des Malleus stylo — hyal cartılage“ (MurıE 1874, p. 188); und „grossierement piritorme, l’oreille moyenne des Sir@niens se retr&cit progressivement vers l’avant et vers l’interieur et se conduit par la trompe d’Eustache, &troit conduit fibreux qui suit le bord posterieur de l’aile interne de l’apophyse pterygoide, puis s’en Ecarte vers l’interieur pour deboucher dans le pharynx“ (ROBINEAU 1965, p. 413). Der pharyngeale Tubeneingang (Abb. 5) zeigt keine Besonderheiten. Das Lumen der Tuba Eustachü ist sehr weit und eigentlich mit dem Begriff „Röhre“ unzureichend charakterisiert. Das Lumen hat vielmehr am etwas über 1 cm langen Abschnitt am pharynzseitigen Anfangsteil der Tuba Eustachii die Form einer hohen Spalte (Abb. 6). Bereits die in Höhe der rostralen sichelförmigen Falte vorhandene Mündung könnte als eigentliche funktionelle Öffnung der Tuba Eustachii in den tympanalen Raum angesehen Zur Anatomie des Gehörorganes der Seekuh (Trichechus manatus L.) U) werden. Morphologisch ist keine eindeutige Entscheidung möglich, wo die Tuba Eustachii endet und die hypotympanale Nebenhöhle des Cavum tympanı beginnt. Um die Verhältnisse weiter zu klären, wurden histologische Schnitte von verschiedenen Abschnitten des tubotympanalen Raumes zwischen Pharynx und Mittelohr angefertigt. Dabei bestätigte sich, daß die Tuba Eustachii von Trichechus sehr stark von den üblichen Verhältnissen der Säugetiere abweicht. Schon in dem beschriebenen, kurzen, pharyngealen Anfangsteil der Tube fehlt ein typischer Tubenknorpel. Im Querschnitt finden sich nur einzelne Knorpelinseln inmittel des sehr kräftigen Bindegewebes. Drüsen treten nur im Bereich der ventralen Biegung auf. Die Lamina propria ist deutlich abgegrenzt. Drei schwache Muskelbündel, die nicht näher identifiziert werden konnten, ziehen an der Lateralseite entlang. Der Knorpel verschwindet nach ungefähr einem Drittel der Strecke zwischen Ostium pharyngeum tubae auditivae und Tympanicum. Im Anschluß bildet dichtes, von respiratorischem Epithel überzogenes Bindegewebe eine ventral zum Tym- panalsack geöffnete, breite, V-törmige Rinne (Abb. 6). Medial des inneren Schenkels der Rinne liegen zahlreiche, tief ins straffe Bindegewebe der Lamina propria eingesenkte tubulo-muköse Drüsen, die verzweigt sind und einen weitlumigen Ausführgang besitzen. Wenige seröse Drüsen befinden sich an der Medialseite des inneren Schenkels. Der M. tensor velı palatini entspringt an der Medialseite eines kräftigen Bandes, das, sich stetig verjüngend, vom rostralen Rand des Anulus tympanıcus zum Processus pterygoideus zieht (Abb. 4, 6). Der schlanke Muskel umschlingt den Hamulus pterygo- ideus und strahlt mit einer sich fächerförmig verbreiternden Endsehne ventrolateral des Ostium pharyngeum tubae auditivae ın den weichen Gaumen ein. Der Muskel hat keine erkennbaren Beziehungen zur Tuba Eustachii. Er wird von Ästen aus dem lateral vorbei- zıehenden N. pterygoideus medialis innerviert. Nach WERNER (1960) ist ein auf das Tympanicum verlagerter Ursprung des Muskels ein sekundärer Zustand. Bei den von Domning (1977, 1978) untersuchten Seekühen entspringt der M. tensor veli palatını mit einer breiten flachen Sehne vom Anulus tympanicus und von der Fossa pterygoidea. Die Muskeln der Gehörknöchelchen Bestätigen kann ich die Beschreibung des M. tensor tympanı und des M. stapedius von Domning (1978). Der M. tensor tympanı entspringt an einem Vorsprung an der caudalen Basis des Processus pterygoideus. Er unterquert die Knochenbrücke zwischen Anulus tympanicus und Tegmen tympani und inseriert mit einer kurzen, kräftigen Endsehne am Processus muscularıs malleı (Abb. 6). RoBINEAU (1964 zit. in SABAN 1968) beschreibt den Muskel als „degenere en tissu fiıbreux chez les Sireniens“. In einer späteren Arbeit bildet RoBINEAU (1969) den M. tensor tympanı als gut ausgebildeten Muskelbauch ab, gibt aber keine Beschreibung. Der M. stapedius ist ein fächerförmiger Muskel, der in einer Grube entspringt, die caudomedial des Sulcus nervi facıalıs liegt und von ROBINEAU (1969) und Domnimng (1978) als Fossa muscularis minor des Perioticum bezeichnet wird. Er legt sich dem ihn innervie- renden N. facialis ventral eng an. Mit einer kurzen Endsehne inseriert er an einem kleinen Vorsprung am Caput des Stapes. Arteria carotis interna und Arteria stapedialis Die A. carotis interna gibt unmittelbar nach Eintritt in die Duplikatur der Tympanalsack- wand die A. stapedialis ab (Abb. 6). Erstere verläuft an der Medialseite der hinteren sichelförmigen Falte zum Foramen lacerum medium, wo das Gefäß zusammen mit dem N. caroticus internus in die Schädelhöhle eintritt. Die A. stapedialis liegt dem Promontorium auf und zieht zum Stapes, wo sie durch das sehr enge Foramen intercrurale hindurchgeht. Der Austritt aus dem For. intercrurale des ohnehin schwachen Gefäßes ist nicht mehr mit 374 M. S. Fischer Sicherheit nachzuweisen. Es kann nicht ausgeschlossen werden, daß es sich um ein nutritives Gefäß für den Stapes oder die ihn umgebende Schleimhaut handelt. Eine übereinstimmende Beschreibung der Verhältnisse dieser beiden Gefäße bei Dugong dugon findet sich bei ROBINEAU (1969). N. facialis und Chorda tympani Der N. facıalis nimmt bei Trichechus manatus denselben Verlauf wie bei Dugong dugon (ROBINEAU 1969). Dem sehr kräftigen N. facıalis legt sich im Hirnraum ein dünner Nerv an, bei dem es sich vermutlich um den N. intermedius handelt. Die ungewöhnliche Stärke des N. facıalıs hängt sicher mit der wohlausgebildeten mimischen Muskulatur, besonders der Schnauzenregion, zusammen. Auffallender Weise tritt der Nerv aus dem Cavum cranıi nicht durch einen verknöcherten primären Facıliskanal aus, sondern durch eine, am mazerierten Schädel gut sichtbare tiefe Incisur zwischen dem oralen Pol des Petrosum und dem Tegmen tympanı. Es tritt somit keine verknöcherte Commissura suprafacialis auf. Hingegen ist bei Seekuhfoeten diese Commissur vorhanden (FREUND 1908; MATTHES 1921; HIRSCHFELDER 1936). Da der N. petrosus major lateral dieser Commissur abgeht, wird sie von MATTHES (1921) mit der inneren Facialiscommissur homologisiert. Eine Commissura suprafacıalıs lateralis tritt diesem Autor zufolge bei den Sirenia nicht auf. Während des Verlaufes in der medialen Wand der Paukenhöhle bleibt der N. facialis jederzeit freı sichtbar; es kommt somit nicht zur Ausbildung einer zweiten Abteilung des Facıaliskanales (van KAmpeEn 1905). Nur anfänglich verläuft der Nerv in einem seichten Sulcus facıalis, der auch bei foetalen Exemplaren beschrieben wird (FREUND 1908; HIRSCHFELDER 1936). Lateral der Articulatio incudostapedia und des M. stapedius zieht der Nerv zur Unterseite der knöchernen Verbindung zwischen Tympanicum und Petrosum. Der N. facialis verläßt die Paukenhöhle durch eine Öffnung (For. stylomasto- ideum primitivum), die ventral und caudal vom Anfangsteil des Zungenbeines, dorsal vom Processus posttympanicus und rostral vom Tympanicum begrenzt wird (Abb. 3, 4, 6). Nach Austritt aus dem Cavum tympani gibt der N. facıalis die Chorda tympanı ab. Durch einen kleinen Kanal ım hinteren Schenkel des Tympanicum gelangt sie in die Paukenhöhle zurück. Im weiteren verläuft sie medial der Basıs des Manubrium mallei (Abb. 6). Sie steht in keiner Beziehung zur Articulatio incudomallearis, und so muß der von FLEISCHER (1971) beobachtete Kanal in diesem Gelenk, den er fragend ın Verbindung mit der Chorda tympanı bringt, einem anderen Zwecke dienen. Lateral und dorsal des Ansatzes des M. tensor tympanı zieht die Corda tympanı am Corpus malleı entlang und verläßt dorsal der breiten vorderen Knochenbrücke zwischen Tympanıcum und Tegmen tympani (Fissura petrotympanıca) das Cavum tympanı. ROBINEAU (1969) weicht ın seiner Beschreibung der Verhältnisse der Chorda tympani bei Dugong dugon insoweit ab, als er proximal keinen Durchtritt durch einen besonderen Kanal und einen kurzen Verlauf der Chorda tympanı auf dem Incus und auf dem M. tensor tympani beobachtete. Der mit dem Tympanicum verbundene Anfangsteil des Zungenbeines entspricht dem Tympanohyale (van KAMPpeEn 1905; ROBINEAU 1969). Bei dem mir vorliegenden Exemplar verbindet sich das Zungenbein mit der erwähnten Knochenbrücke zwischen Tympanıcum und Petrosum, dem angrenzenden Processus posttympanicus und dem Exoccipitale. van Kampen (1905) erwähnt nur die erste und letzte Verbindung. Das Zungenbein ist somit opistnotrematisch (Howes 1896). Diskussion Zunächst soll die eingangs gestellte Frage nach der möglichen Homologie der Tympanal- säcke der Trichechidae mit den Luftsäcken der Procaviidae, Equidae und Tapiridae diskutiert werden. FıscHEr (1986) hat die Morphologie der Luftsäcke (Diverticula tubae Zur Anatomie des Gehörorganes der Seekuh (Trichechus manatus L.) 07 Eustachii) der Schliefer und Pferde beschrieben und mit derjenigen der Tapire verglichen. Bei diesen drei Gruppen stülpt sich der mittlere Abschnitt der häutigen Tube ventral zu voluminösen, luftgefüllten Säcken aus. Diese Tubendivertikel sind grobtopographisch mit dem Tympanalsack der Trichechidae vergleichbar. Während jedoch der Tympanalsack von Trichechus eine ventrale Erweiterung des Cavum tympani ist, wird bei den Procaviidae, Equidae und Tapıridae der membranöse Teil der Tuba Eustachii dilatiert. Die Tuba Eustachii ist hier sowohl an ıhrem pharyngealen Anfangsteil als auch am tympanalen Endstück vor dem Eintritt ın die knöcherne Bulla tympanıca geschlossen. Auch der Verlauf der A. carotis interna ist bei den Procaviidae, Equidae und Tapıridae und den Trichechidae unterschiedlich. Während die Arterie bei den drei erstgenannten Gruppen auf der Medialseite des Luftsackes verläuft, durchquert sie bei Trichechus den ventralen Abschnitt des Cavum tympanı. Bei dem Tympanalsack der Trichechidae und den Tuben- divertikeln der Unpaarhufer und Schliefer handelt es sich dabei nach dem derzeitigen Stand unserer Kenntnis nicht um homologe Strukturen. Eine endgültige Klärung setzt jedoch auch hier ontogenetische Untersuchungen voraus. Die Befunde an Trichechns sprechen nicht gegen die von FISCHER (1986) getroffene Annahme, die Luftsäcke als Synapomorphie der Hyracoidea und Mesaxonıa anzusehen. Ob der Tympanalsack der Seekühe ein Primitivmerkmal ist, wie ROBINEAU (1969) annimmt, scheint fraglich, da, wie eigene Befunde an Elephas indicus zeigen, die Tuba Eustachii der Elefanten eine geschlossene Röhre ist. So kann nicht eindeutig entschieden werden, welche Verhältnisse für den Grundplan der Tethytheria (Sirenia + Proboscidea) anzunehmen sind. Wie im folgenden ausgeführt wird, sind funktionelle Erklärungen denkbar, die den Tympanalsack als spezielle Anpassung an das Hören unter Wasser verständlich machen könnten. Aussagen über die Funktion der Tympanalsäcke sind mangels experimenteller Daten nur ın Analogie zu den Verhältnissen bei anderen aquatischen Gruppen möglich, obwohl die Topographie der Tympanalsäcke bei den Cetacea verschieden von der} jenigen der Sirenia ist. Die in größerer Zahl auftretenden Tympanalsäcke („air-sacs“, „sinuses“) der Cetacea umgeben im Gegensatz zu den Verhältnissen bei Trichechus nahezu das gesamte Petrotympanicum. Sie schieben sich sogar zwischen das Petrotympanicum und die umge- benden Schädelknochen. Nach REYSENBACH DE Haan (1957) öffnet sich die Tuba auditiva tympanalwärts in den „pterygoid sinus“; bei den Mystacoceti verliert die Tube die Verbindung zur Bulla vollständig und endet ın diesem Sinus. Die Funktion der Luftsäcke bei den Walen wird in Zusammenhang mit der Notwendig- keit des Druckausgleiches beim Tauchen in große Tiefen gebracht. Dabei sollen sie, ähnlich wie der Mittelohrraum der Pinnipedia mit Schaum gefüllt sein. Soweit mir bekannt ist, liegen keine Beobachtungen an frischtoten Seekühen über etwaige Füllungen der Tympanalsäcke mit Schaum vor und auch die histologische Untersuchung ergab keinen dem „cavernous tissue“ der Pinnipedia vergleichbaren, auffälligen Befund. D. P. DomnınG teilte mir mit, daß er bei der Sektion von gefrorenen oder fixierten Tieren keine Anhalt- spunkte für eine Füllung des Cavum tympanıi mit Schaum erhielt. Eine weitere Funktion der Luftsäcke der Wale soll die akustische Isolation des Petrotympanıcum sein (FRASER und Purves 1960). Für die Tympanalsäcke der Seekühe ist eine ähnliche Funktion erwägenswert. Der Schall wird beim Übertritt aus dem Wasser oder Gewebe in Luft zu 99,9 % reflektiert. Da der Gesang der Sirenen ein gleichzeitiges Hören erschweren, wenn nicht unmöglich machen würde, wäre es von Vorteil, ventral des Gehörorganes eın Luftkissen auszubreiten, das der Isolation gegen selbsterzeugte Laute dient. Interessanter- weise beobachteten BuLLock et al. (1982) die verhältnismäßig geringste Reaktion, wenn sie ihr Versuchstier von ventral akustisch reizten. Obwohl Trichechus nur in geringe Tiefen taucht, kann der Luftsack darüber hinaus auch für den Druckausgleich bestimmt sein, wofür das Auftreten von lockerem Füllgewebe zwischen Ramus mandibulae und Petro- tympanicum spricht. Eine besondere Schwierigkeit des Hörens unter Wasser ist es, eine Schallquelle zu 376 M. S. Fischer lokalisieren. Da sich der Schall im Wasser fast um das 4,5fache schneller ausbreitet als in der Luft und er ungehindert vom Wasser durch das Gewebe zum Innenohr gelangt, kann beispielsweise der Mensch unter Wasser keine Schallrichtung orten. Für ein aquatisch lebendes Tier sind zwei Lösungen dieses Problems denkbar. Entweder gelingt es, eine zeitliche Verzögerung zwischen dem Auftreffen des Schalls auf das linke und rechte Ohr zu erzeugen, was eine morphologische Lösung auf dem Niveau der Ohrregion erforderte, oder in den zugehörigen Gehirnzentren findet eine Auflösung des fast gleichzeitig auftre- tenden Schalls statt. Der Abstand zwischen den beiden ovalen Fenstern beträgt bei einem uns vorliegenden Exemplar von Trichechus 8,5 cm. Berücksichtigt man den Faktor der Schallausbreitung im Wasser, so ergibt sich ein, einem terrestrischen Säugetier vergleichbarer, funktioneller Abstand von ungefähr 2 cm. Dies ist für kleinere Landsäugetiere kein ungewöhnliches Maß. Auf der Grundlage unserer Kenntnis über die akustischen Zentren im Gehirn der Sirenia ist es unmöglich, Aussagen über spezielle Anpassungen an die Schallortung zu treffen. Das Gehörorgan ist zentralnervös gut repräsentiert, denn im Gehirn Dugong dugon ıst das Corpus geniculatum mediale mächtig entwickelt und auch die Colliculi caudales und die Olive scheinen, bezogen auf das Volumen des Gehirnes, vergrößert zu sein (DEXLER 1913). VERHAART (1972) beschreibt die akustischen Zentren des Hirnstam- mes bei dieser Gattung als einigermaßen normal. Da ich keine morphologischen Besonder- heiten gefunden habe, die auf eine Differenzierung der Schallrezeption hinweisen könnten, könnte eine neuroanatomische Untersuchung unter Umständen weitere Hinweise zur Lösung dieses Problems bringen. Der Mangel an experimentellen Daten zur Schalleitung ım Ohr der Seekühe, erlaubt nur vorsichtige, funktionelle Rückschlüsse zu diesem Problem aus den anatomischen Befun- den. Die Untersuchungen von FLEISCHER (1971) zur Schalleitung wurden am mazerierten, trockenen und vom Schädel isolierten Petrotympanıcum durchgeführt und haben damit nur bedingte Aussagekraft. Damit bleibt auch hier als einzige Möglichkeit, Analogie- schlüsse zu den Verhältnissen bei Cetaceae und Pinnipedia zu ziehen. Doch begegnen wir sofort Schwierigkeiten, denn einerseits ıst die Morphologie des Gehörorganes der Cetacea und Sırenia sehr verschieden, andererseits gibt es mindestens zwei kontroverse Ansätze zur Erklärung der Schalleitung bei Walen. Während eine Schule die Knochenleitung als Grundmechanismus annımmt (Norrıs 1968; McCormick et al. 1970), hält die andere eine Rezeption des Schalles auf dem auch für terrestrische Säugetiere üblichen Weg für wahrscheinlichs (Purves und PILLERI 1983). Genau in der Frage, ob der Schall bei den Seekühen ungehinderten Zutritt zum Innenohr hat, oder ob dieses akustisch gegen den umgebenden Schädel isoliert ist, fehlt die empirische Grundlage. Bei Pinnipedia wird das Innenohr bis auf eine Verbindung zwi- schen Petrotympanicum und Squamosum entkoppelt (RErENnNING 1972). Wale haben die akustische Isolation optimiert, indem sich, wie erwähnt, Luftkissen zwischen das Petro- tympanıcum und die umgebenden Schädelknochen schieben. Bei Delphinidae und Phoca- enidae wird das Petrotympanıcum nur an Ligamenten aufgehängt (ROBINEAU 1969). Am Schädel der Sirenia liegt die fast kugelige Fläche des Petrosum der entsprechenden Aushöhlung des Squamosum eng an. FLEISCHER (1978) nımmt an, daß sich dazwischen eine Knorpelschicht (S. 13) oder „elastic soft tissue“ (S. 53) befindet. Nach experimentellen Befunden von TONNDORF et al. (1966) sind Diskontinuitäten zwischen Knochen keine reflektierenden Filter. Damit könnten die Bedingungen für die Weiterleitung von Schall durch die Schädelknochen zum Innenohr gegeben sein. Sofort stellt sich jedoch die Frage, welche Funktion das Trommelfell und die Gehörknöchelchen mit ıhrer einzigartigen Morphologie besitzen, wenn sie nicht der Schallaufnahme und -leitung dienen. Zum Abschluß der Diskussion über Probleme des Hörens bei Seekühen soll noch kurz auf eine von FLEISCHER (1978) erneut aufgestellte Hypothese eingegangen werden. Schon LitLre (1910) vermutete, daß bei den Walen die Bulla tympanıca ın Schwingung geraten Zur Anatomie des Gehörorganes der Seekuh (Trichechus manatus L.) SUN kann. Nach FLEISCHER ist bei den Sirenia und Cetacea das freischwingende Tympanicum neben Malleus/Incus und Stapes das dritte schalleitende Element. Hierdurch werde die Bandbreite der hörbaren Frequenz erweitert, wovon die Seekühe jedoch keinen Gebrauch machten. Daß diese Annahme ohne experimentelle Prüfung als Tatsache hingestellt wird, wurde schon von Lay (1980) bemängelt. Wie bereits ausgeführt, ist schon die Ansicht, die Gehörknöchelchen hätten beim Hören unter Wasser schalleitende Aufgaben, nicht unum- stritten. Schließlich muß die Annahme eines freischwingenden Tympanicum bei Triche- chus ebentalls in Frage gestellt werden. FLEISCHER (1978, S. 14) schreibt: „Looking first at the Sirenians, we see that the tympanic is still a u-shaped element fused with the periotic at both ends (Fig. 5).“ In Wirklichkeit ist das Tympanıcum nicht an seinen dorsalen Enden, sondern an der Medialseite durch zwei Knochenbrücken mit dem Petrosum verwachsen. Deshalb müssen sich von FLEISCHERS Annahmen abweichende Schwingungsverhältnisse ergeben, wobei besonders zu berücksichtigen ıst, daß eine Achse durch die ventralen Punkte der beiden Verbindungsbrücken noch unterhalb des ventralen Randes des Trommelfelles liegt. Ebenso bleibt bei allen diesen Überlegungen der Einfluß der umgebenden Weichteile unberücksichtigt. Abkürzungen Aaa - Arcus anterior atlantıs, A. cco - A. carotis communis, A. cex - A. carotis externa, A. cin -A. carotis interna, A. max — A. maxillarıs, A. sta - A. stapedialis, C. ary - Cartilago arytaenoidea, C. cri — Cartilago crıcoidea, C. tra — Cartilagines tracheales, Co — Condylus occipitalis, Ct -— Chorda tympanı, D. art — Discus artıcularıs, Dupl. -— Duplikatur der Tympanalsackmembran, Eo — Exoccipi- tale, Epı - Epiglottis, G - Grenze zwischen Pars tensa und Pars flaccıda, G. css - Ganglion cervicale superius nervi sympathicis, h. F. - hintere sichelförmige Falte, Hy — Hypophse, I - Incus, J — Jugale, L. apd - Lig. apıcis dentis, M - Malleus, M. bra -— M. brachiocephalicus, M. cap - M. crico- arytaenoideus posterior, M. cop - M. constrictores pharyngis, M. dig - M. digastricus, M. Ica - M. longus capıtis, M. lco — M. longus colli, M. mas - M. masseter, M. ptl - M. pterygoideus lateralis, M. ptm - M. pterygoideus medialis, M. rcv — M. rectus capitis ventralis, M. stm -— M. sternomastoideus, M. stp -M. stylopharyngeus, M. tty - M. tensor tympanı, M. tvp- M. tensor velı palatini, Mea — Meatus acusticus externus, Mm -— Manubrium mallei, N. cin - N. caroticus internus, N. fac - N. facıalıs, N. las - N. laryngeus superior, N. lin - N. lingualis, N. mas — N. massetericus, N. pm -N. pterygoideus medialis, Oes — Oesophagus, Opta — Ostium pharyngeum tubae auditivae, P - Petrosum, Pm — Palatum molle, R. man — Ramus mandibularıs, R. pha -— Ramus pharyngeus (N. vagus), S - Mediosagittalschnittebene, Sq — Squamosum, Sso — Synchondrosis sphenooccipitalis, St — Stapes, Sty - Stylohyale, T. sym -— Truncus sympathicus, Tt — Tegmen tympanı, Tym - Anulus tympanicus, Tys - Tympanalsack, V. jug - Vena jugularis, v. F. - vordere sichelförmige Falte, IX-N. glossopharyngeus, X - N. vagus, XI -N. accessorius, XII - N. hypoglossus. Danksagung Bei Herrn Prof. Dr. W. MAIER und Herrn Prof. Dr. Dr. D. STArck, beide Frankfurt/M., möchte ich mich ausdrücklich für die vielen Diskussionen und die kritische Durchsicht des Manuskriptes bedanken. Frau Dr. G. KLAuEr, Frankfurt/M. danke ich herzlich für die Diskussion besonders der histologischen Befunde. Das Manuskript wurde kritisch von Frau Dr. E. MıcKoLEit, Tübingen, Herrn Dr. D. P. Domnıng, Washington, und Herrn Dr. G. MIcKoLEIT, Tübingen, gelesen, wofür ich ihnen sehr danke. Der Zeichnerin des Zentrums der Morphologie in Frankfurt, Frau M. Roser, danke ich sehr herzlich für die mit großer Geduld und Können angefertigten Zeichnungen. Ebenso danke ich Frau U. TRAUTMANnN, Zentrum der Morphologie, und Herrn Dipl. Biol. E. Frey, Tübingen, für die photographischen Arbeiten. Frau U. RUITER, Zentrum der Morphologie, danke ich für die Anfertigung der histologischen Schnitte. h Bei Herrn Prot Dr. Dr. D. Starck bedanke ich mich sehr für die Überlassung des wertvollen Materiales. Herrn Dr. R. K. Bonpe vom U.S. Fish and Wildlife Service, Gainesville, Florida, möchte ich an dieser Stelle meinen verbindlichsten Dank für die hilfreiche Zusammenarbeit und Übersendung des juvenilen Tieres aussprechen. Dem Leiter der Zoologischen Schausammlung, Herrn Dr. G. MIcKOoLEIT, danke ich herzlich für die Möglichkeit, daß ich dort die Sektion des Tieres durchführen konnte. 378 M. S. Fischer Zusammenfassung Die Weichteilanatomie des Gehörorganes von Trichechus manatus L. wird beschrieben. An den beiden untersuchten Exemplaren dieser Art endet der äußere Gehörgang in einem Blindsack, der nicht mit dem Trommelfell ın Kontakt steht. Das vom Manubrium mallei zeltartig nach lateral vorgebuch- tete Trommelfell wird von einer ungewöhnlich derben Bindegewebsplatte gebildet. Eine weitere Besonderheit des Mittelohres von Trichechus ist der Tympanalsack. Die eigentliche Paukenhöhle, nämlich der Raum zwischen Tympanalring und Petrosum, ist ventral zu einem geräumigen, membra- nösen Sack erweitert. Er dehnt sich zwischen Schädelbasis, Processus pterygoideus und Anulus tympanicus aus. Der Tympanalsack ist als Teil des Cavum tympanı aufzufassen. Er wird wie dieses von einem mehrreihigen Flimmerepithel ausgekleidet. Die Tuba Eustachii weicht sehr von den üblichen Verhältnissen der Säugetiere ab; so tritt z.B. kein Tubenknorpel auf. Nach dem Vergleich des Tympanalsackes der Trichechidae mit den Luftsäcken (Diverticula tubae Eustachii) der Procavii- dae, Equidae und Tapiridae ist eine Homologie der beiden Strukturen abzulehnen. Vorsichtige Überlegungen zur Funktion der Tympanalsäcke legen nahe, diese als Luftkissen anzusehen, die der Isolation gegen selbsterzeugte Laute dienen. Die Probleme des Richtungshörens unter Wasser und der Schalleitung werden auf die Verhältnisse der Seekühe bezogen und diskutiert. Literatur Burrock, T. H.; Domning, D. P.; Best, R. C. (1980): Evoked brain potentials demonstrate hearing in a manatee (Trichechus inungnis). J. Mammalogy 61, 130-133. BuLLock, T. H.; O’SHEA, T. J.; McCrune, M. C. (1982): Auditory evoked potentials in the West Indian Manatee (Sirenia: Trichechus manatus). J. comp. Physiol. 148, 547-554. Craupıus, M. (1867): Das Gehörorgan von Rhytina stelleri. Mem. Acad. ımp. St. Petersbourg XI (No. 5), 1-12. Dexter, H. (1913): Das Hirn von Halicore dugong Erxl. Morph. Jb. 45, 97-19. Dis, C. (1909): Beiträge zur Kenntnis der Morphologie und postembryonalen Entwicklung des Schädels bei Manatus inungnis Natt. Morph. Jb. 39, 83-145. Domning, D. P. (1977): Observations on the myology of Dugong dugon (Müller). Smithson. Contr. Zool. 226, 1-57. — (1978): The myology of the Amazonian manatee, Trichechus inungnis (Natterer) (Mammalıa, Sirenia). Acta amazon. 8, Supl. 1, 1-81. Doran, A. H. G. (1876): On the morphology of the ossicula auditus. Trans. Linn. Soc. ser. 2 (zool.) 1, 371497. FiscHEr, M. $. (1986): Die Stellung der Schliefer (Hyracoidea) im phylogenetischen System der Eutheria. Cour. Forsch.-Inst. Senckenberg 84, 1-132. FLEISCHER, G. (1971): Über Schwingungsmessungen am Skelett des Mittelohres von Halicore (Sirenia). Z. Säugetierkunde 36, 350-360. — (1973): Studien am Skelett des Gehörorgans der Säugetiere, einschließlich des Menschen. Säugetierk. Mitt., 21, 131-239. — (1976): Über die Verankerung des Stapes im Ohr der Cetacea und Sirenia. Z. Säugetierkunde 41, 304-317. — (1978): Evolutionary priniciples of the mammalıan middle ear. Adv. Anat. Embryol. Cell Biol. 55, 1-70. FRASER, F. C.; Purves, P. E. (1960): Hearing in Cetaceans. Bull. Brit. Mus. (Nat. Hist.) 7, 4-140. FREUND, L. (1908): Beiträge zur Entwicklungsgeschichte des Schädels von Halicore dugong Erzl., Zool. Forschungsreisen in Australien und dem Malayischen Archipel; 4. Bd.: Morphologie verschiedener Wirbeltiere; Semon Zool. Forschungsreisen IV (6); Jena. Denkschr. VII, 41-108 (559-626). HIRSCHFELDER, H. (1936): Das Primordialcranıum von Manatus latirostris. Z. Anat. Entw.-gesch. 106, 497-533. Howes, G. B. (1896): On the mammalian hyoid, with especial reference to that of Lepus, Hyrax and Choloepus. J. Anat. Physiol. 30, 513-526. Hyrtı, J. (1845): Vergleichend -— anatomische Untersuchungen über das innere Gehörorgan des Menschen. Prag: Verlag F. Ehrlich. Kampen, P. N. van (1905): Die Tympanalgegend des Säugetierschädels. Morph. Jb. 34, 321-722. Krauuw, C. J. van DER (1931): Auditory bulla in fossil mammals. Bull. Amer. Mus. nat. Hist. 62, 1-352. Lay, D. M. (1980): Review: Fleischer, G. Evolutionary principles of the mammalian middle ear. J. Mammalogy 61, 388-389. LirLıe, D. G. (1910): Observations on the anatomy and general biology of some members of the larger Cetacea. Proc. zool. Soc. London, 769-792. Zur Anatomie des Gehörorganes der Seekuh (Trichechus manatus L.) 379 MATTHESs, E. (1912): Einige Bemerkungen über das Gehörorgan von Walen und Sirenen. Anat. Anz. 41, 594-599. — (1921): Zur Entwicklung des Kopfskelettes der Sirenen. II Das Primordialcranıum von Halicore dugong. Z. Anat. Entw.-gesch. 60, 1-307. MCcCornMIcK, J. G.; WEVER, E. G.; Pauın, J.; Rınpaway, $. H. (1970): Sound conduction in the dolphin ear. J. Acoust. Soc. Amer. 48, 1418-1428. | Muri, ]. (1874): On the form and structure of the Manatee (Manatus americanus). Trans Zool. Soc. London 8, 127-202. Norris, K. (1968): The evolution of acoustic mechanisms in odontocete cetaceans. In: Evolution and environment. Ed. by E. T. Drake. New Haven: Yale Univ. Press. 297-324. Purves, P. E.; Pırrerı, G. E. (1983): Echolocation in whales and dolphins. London, New York: Academic Press. Repenning, C. A. (1972): Underwater hearing in seals: Functional morphology. In: Functional anatomy of marıne mammals. Ed. by R. J. Harrıson. London, New York: Academic Press. 307-331. REYSENBACH DE Haan, F. W. (1957): Hearing in whales. Acta Oto-Laryng. Suppl. 134, 1-114. RoBInEAU, D. (1965): Les osselets de l’ouie de la Rhytine. Mammalia 29, 412-425. — (1969): Morphologie externe du complexe osseux temporal chez les sireniens. Mem. Mus. Hist. nat. ser A (Zoologie) 60, 1-32. SABAN, R. (1968): Musculature de la tete. In: Trait€ de zoologie, 16 (2). Grasse, P.-P. (ed). Parıs: Masson et cie. 230-471. TONNDORE, J.; OLESENn, M.; Kıng, A. F.; CoTIL£, R. D.; Baker III, D. C. (1966): The effect of osseous discontinuities upon the transmission of vibratory energy across the skull in rats. Acta Oto — Laryng. Suppl. 213, 124-132. VERHAART, W. J. C. (1972): The brain of the seacow Trichechis. Psych. Neurol. Neurochir. 75, 271-232. WERNER, C. F. (1960): Das Gehörorgan der Wirbeltiere und des Menschen. Leipzig: VEB Georg Thieme. Anschrift des Verfassers: Dr. MAarTIın $. FISCHER, Institut für Biologie III, Lehrstuhl Spezielle Zoologie, Universität Tübingen, Auf der Morgenstelle 28, D-7400 Tübingen Z. Säugetierkunde 53 (1988) 380-381 © 1988 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 WISSENSCHAF TEIGTIE RUR ZINN Home range ofan African wildcat, Felis silvestris (Schreber), near Elmenteita, Kenya By T. K. FuLLer, A. R. Bixnevicıus, and P. W. Kar National Museums of Kenya, Nairobi, Kenya Receipt of Ms. 21. 02. 1988 The African wildcat (Fels silvestris; HonAckt et al. 1982) is a common predator in the grasslands and woodlands of East Africa, but almost no quantitative information on its movements ıs available (Kınapon 1977; SMITHERS 1983). Here we describe the home range of a 3.7-kg male wildcat radiocollared near Elmenteita, Kenya (0° 30’ S, 36° 10’E) during August 1937. Temperatures in the area during the study period were warm during the day (13-21 °C) and cool at night (4-10°C); raın was infrequent. Topography is generally flat except for a small, seasonally-dry river bed 5 m deep that bisected the cat’s home range. Habitat types in the area included: Grassland (Themeda triandra, Chloris gayana, Aloe spp., and Withania somnifera), Agrıculture (mostly maize fields), Open woodland (<50 % tree and shrub cover with Acacıa xanthophloea, Euphorbia candelabrum, Opuntia vulgarıs, Senecio petitianus, Cordıa ovalıs, Maerua triphylla, and Warbugıa salutarıs), and Rıverine wood- land (>50 % tree and shrub cover mainly with Ficus sycomorus). The wildcat was captured ın a padded steel foot-hold trap (Victor fox-sızed “Soft- Catch”, Woodstream Corp., Lititz, Pennsylvanıa, USA), then injected via hand-held syringe with a combination of 11 mg/kg ketamine hydrochloride (100 mg/ml; Ketaset, Birstol-Myers Co., Syracuse, New York, USA) and 5 mg/kg promazine hydrochloride (50 mg/ml; Sparine, Wyeth Laboratories, Inc., Philadelphia, Pennsylvanıa, USA). A 40-g radiocollar was fitted around the neck, and a numbered metal tag (3 x 12 mm) was affıxed to each ear. Radiotelemetry locations were determined using a 3-element hand-held yagı antenna, usually while investigators were standing on the roof of a vehicle. The cat was seen once and was located from <50 m away three other times. The other 13 radioloca- tions were determined through triangulation from an average of 2.2 bearings/location made 5 minutes apart, 0.5 km from the anımal. Accuracy was estimated from 13 field tests with transmitters located at positions unknown to one investigator. Mean error of bearings was 6° (range = 0-14°), resulting in a calculated error of #52 m at a distance of 0.5 km. The capture location and 17 telemetry locations were plotted on a topographic map (1:25,000) and recorded as X-Y coordinates on the Universal Transverse Mercator Grid. Home range size was estimated by outlining the minimum perimeter polygon and calculating its area (Opum and KuEnzLer 1955). Changes in signal strength caused by flexion of the antenna on the radiocollar indicated when the wildcat was active. Activity observations were recorded only once within any one hourly interval, sometimes even when a location was not recorded. Habitat distribution was drawn on a topographic map with the aid of aerial photos and ground investigations. Habitat occurrence within the home range was determined by placing a dot-grid over the map and calculating frequency of occurrence. The wildcat was captured on 11 August 1987 and located 17 times during 12-27 August. U.S. Copyright Clearance Center Code Statement: 0044-3468/88/5306-0380 $ 02.50/0 Home range of an African wildcat 381 Radiolocations were obtained more often between sunrise (0700 h) and sunset (1900 h) than at night (n = 11 and 6, respectively). Activity sıgnals (n = 19) indıicated that the cat was mostly nocturnal; all 10 sıgnals obtained during 18000 to 0900 h indicated activity, vs. none of 9 obtained at other times. The wildcat’s home range was estimated as 1.6 km?. Habitat composition of the home range included open woodland (50 %), grassland used for cattle grazing (40 %), riverine woodland (5 %), and agriculture (5 %). The wildcat was located traveling in open grassland during at least 2 different nights, and apparently used different daytime resting sıtes on 8 of 9 days. SMITHERS (1983) stated that Afrıcan wildcats are almost entirely nocturnal, as we also observed. Wildcats in Europe are also active mostly at night, but sometimes cover long distances during the day (Arroıs 1985). Home ranges of wildcats in Europe were reported as 0.5 km? (Nowak and Parapıso 1983), and 0.6-0.7 km? (Leuw 1957 in Kıncnon 1977). Radiocollared European wildcats had daily ranges of 0.3-3.3 km?, and 1.8-12.7 km? over longer periods (Arrtoıs 1985; STAHL et al. 1988). Our estimate of home range size is likely minimal because our marked wıldcat was monitored for only a short time. Wildcats have a wide habitat tolerance, but are conspicuous in open areas (Kınapon 1977) and require cover to rest in during the day (Arroıs 1985; SMITHERS 1983). These requirements were reflected in the mix of habiıtats occurring in the home range of the wildcat we monitored. Acknowledgements We gratefully acknowledge the financial and logistical support of the National Museums of Kenya, the Kenya Museum Society, the Minnesota Department of Natural Resources, B. VAN VALKENBURGH and R. WAyne of University of California, Los Angeles, A. WALKER of The Johns Hopkins University, Baltimore, and P. BEHR of Delamere Estates, Ltd. G. MunGai of the National Museums of Kenya kindly identified plant specimens and D. KuEHn provided helpful comments on the manuscript. References Arroıs, M. (1985): Utilisation de l’espace et du temps chex le renard (Vulpes vulpes) et le chat forestier (Felis sılvestris) en Lorraine. Gibier Faune Sauvage 3, 33-57. Honackı, J. H.; Kınman, K. E.; Koeppr, J. W. (eds.) (1982): Mammal species of the world: a taxonomic and geographic reference. Lawrence, Kansas: Allen Press. Kıncoon, J. (1977): East African mammals. Vol. III A: Carnıvores. New York: Academic Press. Leuw, A. (1957): Die Wildkatze. Merkbl. Niederwildauss. dt. Jagdschutzverb. 16. Nowak, R. M.; PArADIıso, J. L. (1983): Walker’s mammals of the world. Baltimore, Maryland: The Johns Hopkins University Press. Opum, E. P.; KUENZLER, E. J. (1955): Measurement of territory and home range sizes in birds. Auk 72, 128-138. SMITHERS, R. H.N. (1983): The mammals of the southern Afrıcan subregion. Pretoria: University of Pretorıa. STAHL, P.; Artoıs, M.; AUBERT, M. F. A. (1988): Organisation spatiale et desplacements des chats forestiers adultes (Felis silvestris Schreber, 1777) en Lorraine. Rev. Ecol. (Terre Vie), (in press). Authors’ addresses: Topp K. FuLLer, Forest Wildlife Populations and Research Group, Minnesota Department of Natural Resources, Grand Rapıids, Minnesota 55744, USA; AUDRONER. Bıknevicıus, Department of Cell Biology and Anatomy, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA; PiIETER W. Kat, National Museums of Kenya, P. ©. Box 40658, Nairobi, Kenya BEKANNTMACHUNGEN Ausschreibung des FrıTz-FRAnK*-Preises Förderpreis der Deutschen Gesellschaft für Säugetierkunde Die Deutsche Gesellschaft für Säugetierkunde schreibt den Förderpreis in Höhe von 3000,- DM als Anerkennung für hervorragende wissenschaftliche Leistungen junger Forscher aus. Voraussetzung ist eine im Druck vorliegende Arbeit aus den Gebieten Phylogenie und Systematik, Verbreitung, Ethologie, Okologie oder Populationsbiologie der Säugetiere. Die Arbeit muß ın den 3 vorausgehenden Kalenderjahren erschienen sein. Die Autoren dürfen beim Erscheinen der Arbeit nicht älter als 33 Jahre sein. Bewerbungen oder Vorschläge erbitten wır an die Geschäftsstelle der Gesellschaft: Prof. Dr. U. ScHhmiDT, Zoologisches Institut, Poppelsdorfer Schloß, D-5300 Bonn 1, unter Beifügung von 5 Sonderdrucken. Der Jury gehören Wissenschaftler verschiedener Universitäten und Mitglieder der Gesellschaft an. Der Preis wird bei der Eröffnung der Jahresversammlung der Gesellschaft in Lausanne (10.-14. 9. 1989) überreicht. * Unser Mitglied, Herr Dr. FrıTz FRANK, ist am 3. 8. 1988 nach schwerer Krankheit verstorben. Über viele Jahre hinweg galt seine besondere Sorge dem wissenschaftlichen Nachwuchs. 1986 stiftete er den Förderpreis der Deutschen Gesellschaft für Säugetierkunde für die jungen Forscher. Protokoll über die Mitgliederversammlung der Deutschen Gesellschaft für Säugetierkunde e.V. am 3. Oktober 1988 im Vortragssaal des Westfälischen Museums für Naturkunde in Münster Der 1. Vorsitzende, Herr KULZER, eröffnet die Versammlung um 17.30 Uhr. 1. Die Tagesordnung wird angenommen. 2. Herr KuLzer überreicht bei der Eröffnung der 62. Hauptversammlung den Förder- preis der Deutschen Gesellschaft für Säugetierkunde an Herrn Dr. V. SOMMER. Der Vorstand hat beschlossen, den Preis künftig als „FRITZ-FRANK-PREIS — Förder- preis der Deutschen Gesellschaft für Säugetierkunde“ auszuschreiben. 3. Herr ScHmiDT verliest den Bericht über das abgelaufene Geschäftsjahr 1987: Im Berichtsjahr erschien der 52. Band der „Zeitschrift für Säugetierkunde“ in sechs Heften mit 384 Seiten. Zusätzlich wurde allen Mitgliedern das Sonderheft mit den Kurzfassungen der Vorträge und Posterbeiträge der Tagung in Berlin zugesandt. Auf Einladung der Herren Kıös, FRÄDRICH und NIEMITZ fand die 61. Hauptver- sammlung der Gesellschaft vom 28. 9.-2. 10. 1987 in Berlin statt. Über 200 Tagungs- teilnehmer besuchten die wissenschaftlichen Veranstaltungen. Erstmals wurde der Förderpreis verliehen, und zwar an Herrn Dr. MarTIn S. FISCHER. | Durch den Tod verlor die Gesellschaft folgende Mitglieder: Dr. FRITZ FRANK, Echterdingen, Dr. FRIEDRICH KÜHLHORN, München, Prof. Dr. ULrıcHm LEHMANN, Köln. Im Jahr 1987 hatte die Gesellschaft 627 Mitglieder. Bekanntmachungen 383 4. Frau Künnrıch verliest den Kassenbericht für das Geschäftsjahr 1987. 5. Die Kassenprüfer Herr BOHLKENn und Herr SCHLIEMANN haben keinen Anlaß zur Beanstandung gefunden. Sie sprechen die Empfehlung aus, daß die Mitgliederwerbung verstärkt wird und säumige Mitglieder auch die Mahngebühren zahlen müssen. 6. Die Anträge zur Entlastung des Schatzmeisters und des Vorstandes werden einstimmig angenommen. 7. Als Kassenprüfer für das Geschäftsjahr 1988 werden die Herren BOHLKEN und SCHLIEMANN wiedergewählt. 8. Der Vorschlag, die Mitgliedsbeiträge für Vollmitglieder aut DM 95,- zu erhöhen, Beiträge für Studenten (DM 60,-) und Ehegatten ohne Zeitschriftenbezug (DM 10,-) unverändert zu lassen, wird mit drei Enthaltungen angenommen. 9. Mit einer Enthaltung nehmen die Mitglieder die Einladung von Herrn VOGEL an, die 63. Hauptversammlung vom 10.-14. September 1989 ın Lausanne abzuhalten. Als Schwerpunktthemen werden „Wildlife-Management der Säugetiere“, „Endokrinolo- gie und Neurohormone der Säugetiere“ und „Einsatz telemetrischer Methoden in der Säugetierforschung“ gewählt. Die Einladung der Herren SCHRÖPFER und EVERTS, 1990 in Osnabrück zu tagen, wird durch Akklamation angenommen. Für 1991 hat Herr SCHLIEMAnN die Gesellschaft nach Hamburg eingeladen. 10. Der Versammlung wird bekanntgegeben, daß Frau Rasa in das Herausgeberkomitee der „Zeitschrift für Säugetierkunde“ gewählt worden ist. 11. Es wird darüber diskutiert, wie man künftig Poster-Beiträgen zu größerer Wirksam- keit verhelfen kann. 12. Die Sitzung endet um 17.45 Uhr. Prof. Dr. E. Kurzer Prof. Dr. U. SchMIDT Dr. H. FRÄDRICH 1. Vorsitzender Geschäftsführer Schriftführer Internationale Kommission für Zoologische Nomenklatur The following opinions have been published by the International Commission on Zoolo- gical Nomenclature in the Bulletin of Zoological Nomenclature: Vol. 44, parts 3 and 4, 1987 1460 Dasyurus hallucatus Gould, 1842 (Mammalia, Marsupialia): conserved Vol. 45, part 3, September 1988 1516 Taeniolabis Cope, 1882 (Mammalia, Multituberculata): Polymastodon taoensis Cope, 1882 designated as the type species 1517 Viverravus gracilis Marsh, 1872 (Mammalıa, Carnivora): generic and specific names conserved BÜCHBESERESERIINIGEEN GRZIMER, B. (Hrsg.): Grzimeks Enzyklopädie Säugetiere - Band 2. München: Kindler Verlag 1988. 648 S., zahlreiche Abb., Lexikon-Großformat. Leinenausgabe: Subskrip- tionspreis 128,-— DM, später 148,-, ISBN 3-463-42002-3; Luxusausgabe (Halbleder): Subskriptionspreis 168,-, später 198,-, ISBN 3-463-42102-X Bereits kurze Zeit nach dem 4. ist nun Bd. 2 dieser neuen Enzyklopädie erschienen. Darin werden die Arten aus den Ordnungen Scandentia, Primates, Xenarthra und Pholidota behandelt. 29 sachkundige Autoren sind daran beteiligt. Die verschiedenen Kapitel zeigen entsprechend der grundsätzlichen Konzeption des Werkes gewisse Übereinstimmungen im Aufbau, indem sie mit Basisinformationen, einleitenden Kennzeichnungen systematischer Gruppen und tabellarischen Zusammenstellungen mehrerer biologischer Daten versehen sind. Die behandelten Arten werden von den verschiedenen Autoren zwar mit unterschiedlichen Schwerpunkten, jedoch recht ausführlich in ihrer Biologie charakterisiert. Die Tupaias, die neuweltlichen Edentaten und die Schuppentiere sind jeweils von 2 Autoren bearbeitet. Über bereits bekannte Sachverhalte hinaus findet der Leser sehr viele neuere Erkenntnisse integriert, vor allem über physiologische und Verhaltensbesonderheiten, ökologische Ansprüche und rezente Bestandssituationen. Die Bearbeitung dieser artenarmen Ordnungen macht zwar einen ausreichenden, jedoch vergleichsweise geringen Anteil am Umfang des Bandes aus. Der weitaus größte Teil bleibt, wie zu erwarten, den Primaten vorbehalten. Deren Artenvielfalt und Biologie werden entsprechend den Ergebnissen primatologischer Forschung besonders ausführlich behandelt. Informationen über anatomische und physiologische Besonderheiten, über Okologie und Verhalten werden in erzählender und allgemein verständlicher Weise vermittelt, häufig werden soziale Organisationen und Gruppendynamik in besonderer Weise beschrieben. Den in natürlichen Bestän- den stark bedrohten Menschenaften sind besondere Kapitel gewidmet. Zum Teil berichten mehrere Autoren hier aus eigener Erfahrung an Wildpopulationen über die gleiche Art. Probleme von Auswilderungen, in menschlicher Obhut nachgezogener Individuen und Mafßnahmen zum Arten- und Biotopschutz werden an diesen Beispielen besonders eindrucksvoll geschildert. Auch dem Menschen ist ein umfangreicher Teil des Bandes gewidmet, insbesondere seiner Entstehungs- und Ausbreitungsgeschichte, seiner körperlichen Besonderheiten, Vielfalt und basalen Verhaltensweisen. Beeindruckend sind auch in diesem Band die zahlreichen farbigen Situationsbilder von vielen Arten sowie Bildsequenzen von Bewegungsweisen. Dieser Band stellt eine ausgewogene und schöne Zusammenstellung über die heute bekannte Biologie der bearbeiteten Arten dar. Für Laien und Fachleute ist er eine vielfältige Informationsquelle. _ D. Kruska, Kiel Guraya, $.: Biology of Spermatogenesis and Spermatozoa in Mammals. Berlin, Heidelberg, New York, London, Paris, Tokyo: Springer Verlag 1987. 430 pp., 85 fıgs. DM 298,-. ISBN 3-540-17143-6 Unsere Kenntnisse über die Biologie der Spermien und über die Spermatogenese sind in den letzten 20 Jahren auf Grund der Anwendung neuer Techniken, aber auch durch Integration der Ergebnisse verschiedener Disziplinen, enorm angewachsen. Das Schrifttumverzeichnis dieses Buches (53 Seiten) umfaßt, von ganz wenigen Ausnahmen abgesehen, fast nur Arbeiten seit 1960. Es handelt sich um einen umfassenden und detaillierten Bericht unter Berücksichtigung morphologischer (Ultrastruktur), histochemischer, biochemischer, immunbiologischer und physiologischer Aspekte vor allem auf molekularbiologischer Ebene. Da der Autor durch eigene Forschungen erheblich zum heutigen Kenntnisstand beigetragen hat, ist dies Buch weit mehr, als ein Sammelreferat. Zwei Hauptteile, Spermatogenese und Spermatozoa, sind in 12 Kapitel gegliedert. Als Beispiel sei der Abschnitt über SERTOLI-Zellen (20 Seiten) genannt, der eindrucksvoll belegen mag, wie die Einführung neuer Techniken und interdisziplinäres Denken es erreicht haben, daß Hypothesen durch sicheres Wissen ersetzt wurden. Von Interesse für den Säugetierforscher sind Hinweise auf die wesentliche Bedeutung von Zellstrukturen für die Systematik und Phylogenie. Für Veterinäre, Mediziner (Andrologie, Sterilitätsforschung) und Biologen ist das Werk von grundlegender Bedeutung. D. STARcK, Frankfurt/M. Erscheinungsweise und Bezugspreis 1988: 6 Hefte bilden einen Band. Jahresabonnement Inland: 308,- DM zuzüglich 13,80 DM Versandkosten, Gesamtpreis 321,80 DM einschließlich 7% Mehrwert- steuer. Jahresabonnement Ausland: 308,- DM zuzüglich 18,- DM Versandkosten. Das Abonne- ment wird zum Jahresanfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, Spitalerstraße 12, D-2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesellschaft für Säugetier- kunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. Z. Säugetierkunde 53 (1988) 6, 321-384 | TuYı va N N ei Iy { „ ; - 1 # UY 77 $ > £ . > - a a ‘ “ g ı ’ { = \ | j s N D . 1} N, “ | u) i ) ar, .} u ir . = = ee = = = a ss Sao & > 177) NIAN INSTITUTION „DEREN 3 IuvyJg EDEN BRARIES SMITHSONIAN ug u 7 - 7 a D, 5 ut = - = Did. = u. BE X = & e- 4 ‚ EB B yG = < EIN < EI < DEF Ze - = re : dd o = = EN = fe) Bes r z = _ - - z IWS S3IYyVY8I1 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS ° e) = oO s) = = = Ds > = e) 2 > ESGDÄR > = > m >) I GG, ne) = Be) 5 m 2 Th m o = | _ (42) —_ (42) = [7)) ” - NOILNLILSNI NVINOSHLINS S3IUVYUAIT LIBRARIES SMITHSONIAN = 0 zZ el = er wo = < zZ A s „ Fer oO SE. ®, I . oO: = N (02) (02) N % N wo n oO SE ewr L ®) = = E 2 E u > > = > = “ > < (42) = zZ 142) ze u) “ 2 : INS SIIUVYEIT „_MITRSONIAN _ INSTITUTION NOILNLLLSNI NVINOSHLIWS N z z 1 = er ü ul u (88) .s Nez - GG, = < = N : : 7 : = x 2 en # =) = [@) N I = PN = E 2 NIAN_INSTITUTION NOLENLELSNI NNINOSHIINS SS 13V HB BIT EEE nl =) ze ° — (e) 7 ee) = es) S _ [08] = 4,2 5 D N = En) = {> = u > 2 aa = = N B = m, a = a & z | INS _S3 IuV4811 uE BRARI ES SMITHSONIAN INSTITUTION NOILALLLSNI NVINOSHLIWS, zZ < “ zZ < ° 2 5 z | g er z \ = z a ee z 3 BR: 5 E 3 ® IE II © E we EN: = = Sy 2 = z = > = Ben > = N =, = > 177 = 77) Ber 4 (72) NIAN INSTITUTION NOILNLILSNI_ NVINOSHLINS 3 I8V4alıl LIBRARIES GE I < a i . < = < as & = S [6] Eu D&D: ®) = oO _ (e) ix > 3 zZ - = u INS S3lUV4U8I1 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI 3 : 3 = a ie) ER 2 oO PER \ = = 2 ER 7 3 = = Re > ESDÄH > E = Sr 2 EG = - 2 no m 2 Th m 2 m = on = 7 = 7) NOILNLILSNI NVINOSHLINS S3I4UVYSII_LIBRARIES z 72) z (02) — N 02) zZ ee = en z SSO - 2 ) 2 o% ZEN CS = zZ Er Ei Dr z 5 . 3 5 3 5 2 3 LIWNS S31UV4Y911 SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS” N ® z A 2 Me = RX v SR = = "3 = N KL = % = < 3 I N Pr = Ei er a e N cm = Re = ) = I A EN IDIINOSBEINS 2 IuVvyYgı nut BRARI ES SMITHSONIAN_INSTITUTIC 2 z =. z u NN: =. u. ze. BA < E r = < E = = 5 S = = = oO _ (=) = ) ar — oJ r us 4 V4g911 LIBRARIES OB DL EN ROTEILERESIL WININOSBEINIS ni IuVvy4g r an [®) Fee oO Dam UN © De "09 u x — = En 20 S g N 2 = Ey == r > N m a /; = = a NR m 2 4 m 2 m 2 17) = = .) mies TUTION & NOILNLILSNI_ NVINOSHLINS, 53 iuvyg _L iB RAR I En NILSEHIAN \ os: = = NE < N 3 5 m SE 5 = E = E z = ww; > nee LIBRARIES SMITHSONIAN _ NOILNLILSNI _NVINOSHLIWS z | wu = nn = N a [42] = & a u 4 \ N Zus 4 x er N en u en < - NN == c ne = Bi c | 2 2 A a = = ie TUTION 4 NOLLMLILSNI_NVINOSHLIWS „3 Iu3Vv4g8 ll LIBRARI EM. SONAITINSN (e) = e) = 6) u u _ @ = ev) = u 5 2 5 2 = ‚> - = E= = m, in 2 m z m 2 n 2 (9) — | = vy811 zei B RARI ES, DU Blolal „ NVINOSHLIWS „53 IuVy48 e = 5 2 il © = $ ” ? 370° i E 2 = ee 2 z = 2 3 - _NVINOSHLINS S3IUVYSI1_ LIBRARIES _INSTITUTIK 0 20 + Ä zZ = 2 ul 177) Wi 77 : Lu 7 _ 4 | = 4 = 0-4 2 3 NN < = = a : Ss 885 = 5 = = ) = =) = fe) er = ) 2 _ > | SMITHSONIAN INSTITUTION NOILNLILSNI s31yvuyg = 5 Er an ae = | -& wu "0 nn [80 Di 2 5, ® 5 D 5 > rs > > > r Bu) we’ &o m Pr) = za == €. 2) == D zei. z n z NOILNLILSNI NVINOSHLINS 53 IuVy8 ul, y 2 = < so < Ss < 3 z | z 4 zZ 2. O Be 2 a 7 5 es oO, „ ®) T = = 2, = = = me: ie : u LIBRARI ES SMITHSONIAN _ INSTITUTION NOILNLILSNI_ = 5 - 5 a 77 Lu \ = De. = & = & <<, < < EZ 3 u) EU EA 5 rs.“ Fusyn a ee Ze ee Ze TE T er eer T er Ve Pi m. w% F # BE i IE} BER yeuanea erg ni wage Ka R » nm vom Er Le I Al 2 ve” h ” TEN 0 My auf ai ven: INH N, a En sap EA wars ra nd Es ya: Dean age onen nun INN] 3 9083 0095 A I