Age variation of the third upper molar in Eothenomys smithii-香川大学学術情報リポジトリ

〟α椚椚αJ5ね‘め21：1−13（1996） ⑥theMammalogicalSocietyofJapan

Agevariationofthethirduppermolarin

且bi励e几Om〝g gm王統ff

YukibumiKANEKO 凡−（｝／（融・‘イ／ム止≠正一り・．品（・J小・‘1J■Jご‘れ、‘J／／州．ん桝冊【㌻′∼′nl叫、・．れ血川（7／J′′1丁（し力外川 Fax，0878−36−1652

Abstract．A study was made of age variationin the size and

enamelpatternsofthethirduppermolarof99Eothenoクク砂SSmithii

specimensfromJapan．Nosignificantagevariationwasfoundin

eitherthefrequency ofthepatterns，Or thewidth ofthe dentine

COnfluentspacebetweenthesecondandthethirdtriapgles・Deep

lingualreentrantfolds，Ontheposteriorloop，appearlnSpeCimens where the condylobasallength（CBL）is of22−24mm，then the

patternwith a shallowreentrantfoldincreasesin frequencyin

largerCBLclasses．Thedepthoftheinnerfoldshowedthesame

tendency as the changesin the patterns．A significant associa− tion，however，between five enamelpatterns and age classes，

dependsonclassificationaccordingtoCBLorbodyweight．This

provedinsignificantinfiveCBLclasses，butsignificantinthree

CBL orbody weight classes．A gradualtransitionin the age

variation of the posteriorloop patterns was found among Eo−

thenon砂S SpeCies which have rootless molars throughoutlife．

The simple enamelpattern form slgnificantlylnCreaSedin fre−

quency with advancing agein E．andeYSOniand E．shanseius， resemblingClethrionoクク砂Sgla柁OlusandC．n4舟canus；Ontheother hand，inEプ聯Ius，E．inez，E．eva，E．chinenめE．wardi，E．custos andE．proditornoagevariationwasfoundontheposteriorloop， thusresemblingMicYVtuSPennsylvanicus．E．smithiishowsalittle

agevariationintheenamelpatterns，thevariationofwhichisof

anintermediatetype． Keywords：ageVariation，enamelpattern，EothenonサSSmithii，Sizeofmolar， thirduppermolar． WithregardtothephylogenyoftheArvicolidae，BauchauandChaline（1987）， andChalineandGraf（1988），COnSideredthat，basedonacomparisonofmolar

structures，theocclusalenamelpatternsofthethirduppermolartendedtovary

fromsimpletomorecomplexforms．ThegenusClethriononv）Sdevelopsmolar rootswith advancing age，Whereasthe genusEothenonys develops no roots． Thetwogenera，however，reSembleeachotherinmanyothercharactersofthe skulland dentalmorphology（Hinton1926，Kaneko1990，1992），andintheir karyotypes（Yoshida et al．1989）．Throughthe ontogenetic process of C・ glanolus（Zejda1960），C．ndbcanus（Abe1982）andE．andeYSOni（Miyao1966， Kitahara1995），alargeproportionofmolarschanges from complex enamel

肋∽研α／Sf〟め′ 21：1996

PatternStOSimplerforms．Noagevariationwasfound，however，Onthesame molarinE．smithii（Tanaka1971）．Tanaka，s（1971）resultsf6rE，Smithiimay

havebeenbiasedbecause ofhisrelativelysmallsample group ofspecimens COllectedduringjustoneperiodoftheyear，Whenfullyadultanimalsmayhave

been absent．

Thepurposeofthisstudy，therefore，WaStOreeXaminetheagevariationin bothsizeandenamelpattern ofthethirduppermolars ofE．smithii，andto

COmparetheresultswiththoseofotherEothenonwsspecies．

MATERIAIJSAND］Ⅵ：ETHOS

Atotalof99specimensofE．smithiiwerecollectedatMinoura，Toyohama District，Kagawa Prefecture，Japan，（34002′30′′N，133037′30”E）．Specimensin

each ofthe12monthsweresampled at one period duringtheyears1977−80

（Kaneko1989）．Thecollectingsite for this studywaslessthan50kmfrom Tanaka’s（1971）siteonthesameisland，Shikoku．Fivemeasurementsofthe

thirduppermolarandthecondylobasallength（CBL）weretakenfromcleaned

Sku11s，these were：tOtallength（TM3L），anteriorlength（AM3L），pOSterior length（PM3L），the width of dentine confluent spaces between the first and

SeCOndtriangles（WDC）andthedepthofthethirdlingualreentrantfoldorthe

posteriorroop（DRF，Fig．1）．Toothdimensionsweremeasuredtothenearest 朋界雷一・幣・ヾ小︶ 舶摘㌧笥即Y小．・．J

〟境野じ

桑票し

摺

用

只野．． 只京．小〃卜＝》

（）ハ嘲

Fig．1．Measurements taken（1eft），and enamelpattems（AO−A3and Pl−P4）on the third upper molar according to Tanaka（1971）．TM3L，AM3L，PM3L，WDC and DRF are ex−

Kaneko，Age ua7iation〆the molarinEothenomyssmithii

0．01mmusingastereo−microscope（Nikon，SMZ−10）withanobjectivemicro− meter（Kogaku，minimuminterval＝0．05mm）．TheCBLwasmeasuredtothe nearestO．1mmwithadialcaliper（minimuminterval＝0．05mm）．

Tanaka（1971）definedtheenamelpatternsformedontheocclusalsurface

bytheenamellamellae onlyin figures；however，inthis study more precise criteria havebeen used．Four patterns（AO−A3）in the shape ofthe dentine

COnfluentspacesbetweenthesecondandthirdtriangleswererecognized．In AO，thelamellaedonotformtwotriangles，butawidedentineconfluentspace instead．AlandA2areintermediatepatternsbetweenAOandA3（AIshows asmallerprotrusionoftheenamellamella，andA2alargerprotrusion）．InA3 thelamellaeformtwocompletetriangles．Fourotherpatterns（Pl−P4）were Observedontheposteriorloop，OrOnthefourthsalientangle．InPlthepattern iscomplex，Withthreereentrantfoldsonthelingualside，Withthethirdfold

exceedingthetransverseline atthe anterior edge ofthesalientangle ofthe

POSteriorloop．P2isintermediatebetweenpatternsPlandP3．P3hasthree Salientangleswithastraight−Sidedposteriorlooponthelingualside．InP4 thepatternissimplewithtworeentrantfoldsonthelingualsideandwithout COnCaVityontheposteriorloop．Enamelpatternswereobservedontheright Orleftmolarunderastereo−microscopewitha X201ens． Inthisstudy，CBLwasusedasanapproximateindicatorofage，becauseit COrrelatespositivelywithageasdefinedbyrootdevelopmentinClethrio

ndbcanus（Kaneko1990）．As there have been no reportsindicating sexual

differencesin either size or enamelpatterns，both males and females were COmbinedforanalysis．

RESULTS

AsCBLincreased，bothtotallength（TM3L）andanteriorlength（AM3L） increasedsignificantly（r＝0．661，P＜0．001andd．f＝97inTM3L；r＝0．676，P＜ 0．001，d．f＝97in AM3L）．As CBLincreased，the posteriorlength（PM3L） increased untilCBL reached22．5mm whereit reached asymptote，thougha

significant regression coefficient was calculated throughout the size of CBL

（γ＝0．305，0．001＜♪＜0．01，d．′＝97，Fig．2）．

The depth ofthe reentrant fold ofthe posteriorloop（DRF）was nearly

COnStantagainstCBL＝22−25mmanddecreasedslightlyinCBLslargerthan25

mm．WiththeincreaseofCBL，Widthofthedentineconfluentspaces（WDC） remained almost constant．Regression coefficients between CBL and DRF， and between CBL and WDC wereinsignificant（r＝−0．198，n．S．，d．f＝97in DRF；r＝−0．057，n．S．，d．f＝97inWDC，Fig．3）．

Theaveragelength（X），Standarddeviation（SD），andcoefficientofvaria− tion（CV）ofthesefivemeasurementsaretabulatedforfivesizeclassesofCBL．

AverageTM3LandAM3LincreasedcontinuouslyasCBLincreased，Whereas

PM3L ceased toincrease from the CBL＝22．O mm class onwards．Thecoeffi−

CientofvariationisgreaterinPM3LthaninTM3LandAM3Lforeachsize

肋椚桝αJSf〟め′ 21：1996 00 0 008 0 00（） 0 ◎ O ′． （〉 血 TM3L 0 0 0 g 茸。‰冨 ∴、∼ざ、｝・ 1．5 0 0 0 AM3L 0 0 ◎‰○ 0 0 ∵∵叫． ◎ 卵S 。。 00 ＆0 0g 0︵ 。 亀。。 ○ ○ ▼ 000 VOβ 0 0 β 〆叩。。。

00 0 0

0 0

0 00

PM3L 0 0 0 0 0

000000

さ＿ 0 0 21 22 23 24 25 26 C B L（mm） Fig．2．PlotsofTM3L，AM3LandPM3Lagainstthecondylobasallength（CBL）． ◎ 0 l ㍉．∵ 亀 ㌍◎0 0‰縞。。 0恥○ ◎ 0 こ豆モ土ミ・一二隠

21 22 23 払 25 26

C B L（m） Fig．3．PlotsofWDCandDRFagainstthecondylobasallength（CBL）．

Kaneko，Age variation d the mohlrinEothenomyssmithii

WDCwasslightlylongerinthe20mmCBLclass．Thecoefficientofvariation

wasgreaterwasDRFinthe25mmCBLclass（Tablel）．

The association of enamelpatterns betweenleftand right third upper

molarswastestedusingtheGLteStWithWilliams’adjustment（SokalandRohlf 1973）．Noindependence was shown between right andleft molarsin the patternsofthedentinespacesbetweenthesecondandthirdtriangles（AO−A3； Ga。j＝35．66，P＜0．005，d．f＝9，Table2），Orbetweenrightandleftpatternsofthe posteriorloop（Pl−P4；Ga。5＝43．88，P＜0．005，d．f＝9，Table2）・Similarlyno

associationwasshownbetweenthepatternsofthedentineconfluentspacesand

thoseoftheposteriorloop（AO−A3andPl−P4，Gadj＝6．56，n．S．，d．f＝9，Table3）・ Consequently，theleftthirduppermolarwasusedforfurtherstudies・

In an analysis of the width of dentine spaces（WDC）and the enamel

patterns（AO−A3）betweenthesecondandthirdtrianglesforeachofthefivesize classesofCBL，theO．05−0．10mmWDCclassandpatternA2predominatedin everyCBLsizeclassover22．Omm，andaverageWDCremainedalmostcon− stantthroughoutthesizeclasses（Fig．4）．Noassociationwasfoundbetweenthe enamelpatternsandthefiveCBLsizeclasses（Ga。j＝18．73，n．Sリd．f＝12，Table 4）．Whenthefrequency ofthe patternswasdividedinto threebodyweight classes，aSdefinedbyTanaka（1971），OrbyCBL（20．5一，23．0−，and 25．0−mm Tablel．FivemeasurementsofthethirduppermolarsofEothenonwssmithii． Five CBL classes（mm） 20．5− 22．0− 23．0− 24．0− 25．0− 5 19 21 37 17 1．508 1，720 1．769 0．150 0．082 0．119 9．946 4．757 6．661 1．37 1．50 1．55 1，75 1．86 1．99 1．832 1．908 0．093 0．102 5．064 5．368 1．63 1．76 2．00 2．08 X 孟Ⅳ仙Ma L l h 3 ︶ 1．190 1．215 0．080 0．087 6．679 7．170 1．04 1．03 1．36 1．41

蔓

r ︶ 若槻仙 0．974 1，068 1．122 0．060 0．060 0．073 6．181 5．572 6．534 0．93 0，97 1．00 1．08 1．20 1．25 r _、∴二 O ︶

∵

0．739 0．752 0，076 0．087 10．303 11．510 0．56 0．60 0．90 0．93 0．564 0，730 0．748 0．144 0．071 0．073 25．497 9．760 9．818 0．43 0．57 0．61 0．80 0．88 0．90 Confluent X O．094 0．050 0．064 0．056 0．059 width SD O．026 0．028 0．029 0．032 0．032 （WDC） CV 27．766 55．800 44．479 57．271 54．422 （mm） Min． 0．05 0．00 0．00 0．00 0．00 Max． 0．12 0．10 0．11 0．12 0．10 Depth of X O．170 0．223 0．221 0．217 0．130 reentrant SD O．082 0．076 0．088 0．083 0．087 fold CV■ 48．177 34．187 39．638 38．199 67．154 （DRF） Min． 0．10 0．07 0．06 0．05 0．00 （mm） Max． 0．30 0．40 0．36 0．38 0．30

肋∽∽αJ5′〟め′ 21：1996

廿

25‘）− N＝17 2ん0− N＝37

qlQ2mm

WDC

CBL

（m巾） Fig．4．FrequencydistributionsofWDCmeasurements andenamelpatternsAO−A30fthe dentinespaceforfiveCBLclasses．Aclosedtriangleshowstheaverage，andSDindicates thestandarddeviationoftheaverage．

Kaneko，Age variation〆the mohlrinEothenomyssmithii

258−

N＝17

︼ 7

∵晒

ーーl■二■ニー二云

●

−

慧

20．5−

N＝5

CBL

（mm）

GIG2 α3（払mm

DRF

Fig．5．FrequencydistributionsofDRFmeasurementsandenamelpatternsPトP4forfive CBLclasses．Aclosedtriangleshowstheaverage，andSDindicatesthestandarddeviation Oftheaverage．

ルね∽∽αJ5蝕め′ 21：1996

Table2．A test ofindependence for frequencies of the enamel

patterns（AO−A3andPトP4）betweentherightandleftmolarsof g（7〃圧りれり町5封〃〟／㌦． The left Confluentpatternsofthe2ndand3rdspaces AO AI A2 A3 Total Theright a t O 1 2 3 0 A A A A T 3 4 9 3 9 1 6 1 9 2 1 6 0 9 1 1 0 2 9 1 2 5 6 0 0 4 2 6 1 1 Patternsofposteriorloop PI P2 P3 P4 Theright PI P2 P3 P4 Total 2 9 5 3 9 4 3 1 9 0 5 2 1 8 1 1 4 0 0 0 4 3 1 5 7 4 3 0 4 3 2

Table3．A test ofindependence for frequencies between the

enamelpatterns（AOLA3andPl←P4）ofEothenonwssmithii． Confluentpatternsbetweenthe2ndand3rdspaces AO AI A2 A3 Total Posterior Pl 1 5 loop P2 0 7 patterns P3 1 6 P4 0 1 Tota1 2 19 8 4 9 1 2 6 2 2 0 3 2 1 6 1 1 4 3 1 9 4 4 8 3 9

based on Table4），nO Significant association was shown between the two

dimensions（Gadj＝8．13，n．S．，d．f＝9forCBL；Ga。j＝4．83，n．S．，d．f＝6forbody weight，Table5）．Thus，thevariationofthepatternofthedentineconfluent SpaCeS（AO−A3）isindependentofage． Inananalysisofthedepthofthereentrantfold（DRF）andoftheenamel patternsoftheposteriorloop（Pl−P4）foreachCBLsizeclass，patternSPl−P3

appearedwithsimilarfrequenciesinCBLclassesfrom22mmto24mm，aSdid

DRF，Withalmostthesameaverage，thoughwithinawiderangebetweenO．05 andOA5mm（Fig．5）．Inthe25mmCBLclass，aVerageDRFbecameslightly Shallower，andpatternP4appearedforthefirsttime，andP2becamethemost

frequent pattern，No associ？tion was fらund between the patterns of the

posteriorloopandthefiveCBLsizeclassesatthe5％1evel（Gadj＝19．36，0．05＜

P＜0．1，d．j：＝12，Table4）．However，Whenthepatternsoftheposteriorloop wereclassifiedintothreebodyweightclasses，aSdefinedbyTanaka（1971），and

助neko，Age yaYiation qf the molarinEothenomyssmithii

Table4．A test ofindependence for frequencies between the

enamelpatterns（AO−A3and Pl−P4）andfive CBLsize classes of

励〃zβ乃0クワ砂ざ5研グfカグブ． Confluentpatternsbetweenthe2ndand3rdspaces AO AI A2 A3 Total Five 20．5− CBL 22．0− classes 23．0一 （mm） 24．0− 25．0 Total 0 3 3 8 2 6 1 3 ハU 4 0 2 9 1 1 1 5 4 9 3 2 ﹁⊥ l l 1 6 5 ∩プ 1 7 7 9 1 2 3 1 9 Posteriorlooppatterns PI P2 P3 4 P O T t a Five 20．5− CBL 22．0− classes 23．0一 （mm） 24．0− 25．0− Total 1 3 1 7 2 4 1 1 1 4 2 4 7 3 8 4 1 3 2 2 3 7 4 8 1 5 9 1 7 7 9 1 2 3 1 9

Table5．A test ofindependence for frequencies between the

enamelpatterns（AOuA3andPl−P4）andthreebodyweightclasses Orムー〃れリハり町・ゞ封〃／仙／∴ ConfluerltpatternSbetweenthe2ndand3rdspaces AO AI A2 A3 Total 3 2 4 9 1 1 2 1 3 6 1 1 9 5 5 9 6 2 9 3 1 8 2 4 1 6 Posteriorlooppatterns PI P2 P3 a t O T 4 P Body lO．0− Weight 20．0− （g） 30．0− Total 9 5 5 9 6 2 9 3 6 5 4 3 4 3 1 0 4 2 1 3 3 8 7 8 1 byCBL（20．5−，23．0−and25．0−mmbased onTable4），Significantassociations Werefoundforbothdimensions（Gadj＝17．69，P＜0．01，d．f＝6，forCBL；Ga。j＝ 15．34，P＜0．025，d．f＝6，forbodyweight，Table5）． Whenthetotal（TM3L），anterior（AM3L），andposterior（PM3L）1engthsof

thethirduppermolarwereexaminedinrelationtothefourenamelpatternson

theposteriorloop（Pl−P4）foreachofthefourCBLsizeclasses（Fig．6），itwas foundthataveragePM3LstendtodecreasefromPltoP4atthe24and25mm CBLclasses，WhereasaverageAM3Lincreasedslightly．

10 肋∽∽αJ5′〃め′ 21：1996

0

等・こ＝

CB◎ βド巾岬噛 P−Tb CBヒ25．0− ○ mm mm 1．0 2．1 P・⊥Ⅶ・b CBlj払．0− ‘ ﹂M三＜ 」 M ∑ ‘Ll．1 CBL芦23．0− 0 CBし；受3．0− CBL戎3（ト だギb

。堰。

。握。

CBL言22β− か 2 つ︻ ヒ B C CBヒ22．0− 0 8馳 _P十P βホb 8 1．7 0 1 l PIP2 P3 P4 PIP2 P3 P4 P1 P2 P3 P4 Fig・6・MeasurementsofTM3L，AM3LandPM3LagainstenamelpatternsPl−P4forfour CBLclasses．Aclosedtriangleshowstheaverage，andaverticallineindicatesthestandard deviationoftheaverage．

11 Kaneko，Age variaiion〆themolarinEothenomyssmithii

DISCUSSION

InrevisingthetaxonomicpositionofEothenoクク砂S Smithii，Tanaka（1971） firstshowedthatno agevariationwas found onthe enamelpatterns・Ofthe

dentine confluence between the second and third triangles and the enamel

patternsofposteriorlooponthethirduppermolar．Sixty−Sixspecimensused byTanaka（1971）werecollectedattheendofJulyat1700monMt．Tsurugi， TokushimaPrefecture，Shikoku，Japan．Thebreedingseasonofthisvoleisat itspeakduringJulywhenTanaka（1971）collectedspecimens，andfullyadult VOlesrepresentedonlyasmallproportionofthepopulation（KanekoandMorii 1976）．Inthisstudy99specimenscollectedthroughoutayearwereexamined，

andtheseincludedalargerproportion（26％）ofold adultswithbodyweight

heavierthan30g，thanin Tanaka，s（1971）sample（16％）．Inthe specimens COllectedforthisstudy，however，theposteriorloopofthethirdmolarshowed SOme age Variation，aS Shown by theincrease of the frequency of P2，the appearenceofP4（thesimpleform）andtheshallowerreentrantfold（DRF）in thelargestCBLsizeclass（Figs．3and5，andTablel）．

Table6．Therelationshipbetweenenamelpatternsandskullsizesinspeciesof励theno〝り′S

（datafromKaneko1990，1992）．

Posteriorlooppatterns＊ Species

Size（mm） Type6 TypeP2 TypeP3 TypeP4

E．ざゐα乃5g才〟ざ 12．2− （トM3） 13．卜 Ga。j＝52．02 14．0− ♂．′＝12 15．2− ♪＜0．05 16．1− Total 0 2 6 2 ﹁⊥ 1 1 1 3 9 8 亡U 7 2 0 2 3 6 0 1 1 1 3 5 6 1 0 ハU 2 1 6 3 9 6 7 1 1 1 2 1 8 且オ乃gZ ll．3← （トM3） 12．2− Ga。j＝6．28 13．ト d′＝9，乃．∫． 14．0− Total 0 2 8 2 2 1 0 4 2 3 9 1 1 1 7 0 5 3 2 3 丘、．e〃α 11．3− （Ⅰ−M3） 12．2− Gadj＝5．74 13．ト d′＝6，邦．∫． 14．0− Total 0 2 9 0 1 1 3 2 0 1⊥ 6 1 2 3 3 7 9 1⊥ 0 1 2 5 且ク聯J〟S 21．0− （CBL） 22．0− G8dj＝8．94 23．0− d′＝15，乃．ざ． 24，0− 25．0 26．0− Total 2 3 5 6 5 6 7 2 3 4 6 00 6 8 5 3 ＊Type6hasthreesalientanglesonthelingualside，aShortposteriorloop，andaconfluent dentalisthmusbetweentriangles．ExceptforType6，alltypesappearinginKaneko（1990，・ 1992）werefollowedinthepresentclassification． n．s．：nOn−Slgnificant．

12 肋∽∽αJS≠〟め′ 21：1996

Astatisticallyslgnificantassociationwasfoundbetweentheposteriorloop

patterns（Pl−P4）andthree CBLclasses（20．5−，23．0−and25．0−mm，Table4）， becausetheDRFwasnearlyconstantthroughoutthe22r24mmCBLclassesbut decreasedonlyinthelargest25mmCBLclass（Fig．3aムdTablel）．Further− more，ateStbetweenpatternsPl−P4andthethreebodyweightclasseswould besignificant（Table5），becausebodyweightcorrelatessiginificantlywithCBL （r＝0．911，P＜0．001，d．f＝97）andindividualswithaCBLofmorethan25mm COrreSpOndwiththoseofbodyweightsofover30g．

Asageincreases，thepatternoftheposteriorlooptendstobecomesimple

inClethriononvLSglareolus（Zejda1960）andC．nLfbcanus（Abe1982），Whichhave

rootedmolarsin olderindividuals．Thesimpleformincreasesinfrequency

from the root ratio exceeding63％in C．glayeolus and32％in C．ク頑canus

（Zejda1960，Abe1982）．Duetothelossofthethirdreentrantfoldwithage，the

proportionofthe simple formincreasesin C．gh17mlus and C．ndbcanus．In COntraSt，ageVariationwasnotobservedontheloopinMicrotuspennsyluanicus， Whichremainsrootlessthroughoutlife（Oppenheimer1965）． AmongtenspeciesofEothenon砂Shavingrootlessmolarsthroughoutlife， agradualtransitionisfoundintheagevariationoftheposteriorlooppattern： inE．andeYSOnitheratioofthesimpleformoftheposteriorloopincreaseswith advancingage（Miyao1966，Kitahara1995），andinE．shanseiustheproportion OfthesimpleformincreaseswithincreasingCBL，thoughsamplesfromdiffer− entpopulationswerepooled（Table6），reSemblinginthisrespectClethriononws glareolusandC．rt4bcanus．Incontrast，ageVariationhasnotbeenobservedin

either Einez，eVa，プ聯Ius，Chinenめ，u）aYdi，CuStOS，Or Pnditor，they thus resemble Micntuspenn即Iuanicus，thoughsamplesfromdifferentpopulations WerepOOled（Table6andunpublisheddata）．Thus，E．smithiiisanintermedi− atetypebetweenthesetwogroupsshowingalittleagevariation． InE．smithii，theposteriorlengthofthethirduppermolar（PM3L）ceased togrowwithage，thoughboththetotalmolarlength（TM3L）andtheanterior length（AM3L）increasedwithage（TablelandFig．2）．ThegrowthofTM3L， therefore，isrelatedtothatoftheanteriorlength．However，WhenPM3Lwas plottedagainstthefourposteriorlooppatterns（Pl−P4，Fig．6），theposterior lengthtendedtobecomerelativelyshorterfromPltoP4asCBLincreases．In C，ndbcanus，thesimpleform（P4）increasesgreatlyinfrequency（Abe1982），aS thelengthdecreaseswithadvancingage（Abe1973）．Itissuggested，therefore， thattheincreaseinthefrequencyofP4isduetoashorteningofPM3Lwith

advancing age．Furthermore，in E．smithiiit appears that PM3L does not

decreaseprominentlywithadvancingage（Fig．2），becausetheincrea？einthe frequencyofthesimpleP4patternisrelativelylowerthanin C．rゆcanus． Acknou）ledgements：MyspecialthanksareduetoDr．K．Maeda，NaraUniver−

SityofEducation，forreadingadraft，andtoanonymousrefereesforreading andimprovingthefinalmanuscript．

13 Kaneko，AgevaYhltion d themohrinEothenomyssmithii REFERENCI∃S Abe，H．1973．GrowthanddevelopmentintwoformsofClethriononvLS．II．Toothcharacters，With specialreferencetophylogeneticrelationships．］．Fac．Agr．HokkaidoUniv．57：229−254． Abe，H．1982．Ageandseasonalvariationsofmolarpatternsinaredrbackedvolepopulation・J． Mammal．SocJapan9：9−13． Bauchau，Ⅴ．andJ．Chaline．1987．Variabilitedelatroisi巨memolairesup6rieuredeClethrio ghlYeOlus（Arvicolidae，Rodentia）etsasignificationevolutive．Mammalia51：587−598−

Chaline，J．andJ．−D．Graf．1988．Phylogeny of the Arvicolidae（Rodentia）：biochemicaland paleontologicalevidence，J．Mammal．69：22−33．

Hinton，M．A．C．1926．Monograph oftheVoles and Lemmings（Microtinae），Living and Extinct． BritishMuseum（NaturalHistory），London，488pp． Kaneko，Y．1989．SeasonalchangesofthenumbercollectedandreproductioninEoihenonwssmithii atthefootofalowermountain，Minoura，KagawaPrefecture，Japan．KagawaSeibutsu（15／ 16）：67−74（inJapanesewithEnglishabstract）． Kaneko，Y．1990．IdentificationandsomemorphologicalcharactersofClethriononwsndbcanusand Eothenonws7％ulusfromUSSR，nOrtheastChina，andKoreaincomparisonwithC．ndbcanus fromFinland．J．Mammal，Soc．Japan14：129−148．

Kaneko，Y．1992．Identification and morphologicalcharacteristics of ClethYionomys rtdbcanus， Eotheno〝砂SShanseius，E．inezandEevafromtheUSSR，Mongolia，andnorthernandcentral China．J．Mammal．Soc．Japan16：71−95．

Kaneko，Y．and R．Morii．1976．Altitudinalsurvey of smallrodentsin Mt．Tsurugi，Tokushima Prefecture．Mem．Fac．Educ．，KagawaUniv．II26：43−52（inJapanesewithEnglishabstract）． Kitahara，E．1995．TaxonomicstatusofAnderson’sredMbackedvoleontheKiiPeninsula，Japan，

basedonskullanddentalcharacters．J．Mammal．Soc．Japan20：9−28．

Miyao，T．1966．SmallmammalsonMt．YatsugatakeinHonshu．Ⅴ．Variation ofthe third upper molar patternin ClethrionoタγりJS andersoni．J．Growth5：7−12（inJapanesewith English

Summary）．

Oppenheimer，J．R．1965．Molarcusppatternvariationsandtheirinterrelationshipsinthemeadow vole，MicYVtuSP．Penn即Ivanicus（Ord）．Am．Midl．Nat．74：39，49．

Sokal，R．R．andJ．Rohlf．1973．Introduction to Biostatistics．W．H．Freeman，San Franciso and London，368pp． Tanaka，R．1971．Aresearchintovariationinmolarandexternalfeaturesamongapopulationof theSmith’sred−backedvoleforelucidationofitssystematicrank．JapaneseJ，Zool．16：163− 176． Yoshida，Ⅰ．，Y．ObaraandN．Matsuoka．1989．Phylogeneticrelationshipsamongseventaxaofthe JapaneseMicrotinevolesrevealedbykaryologicalandbiochemicaltechniques．Zool．Science 6：409−420．

Zejda，］．1960．Theinfluenceofageontheformation ofthethird uppermolarinthebank vole Clethriononwsglareolus（Schreber，1780）（Mamm．：Rodentia）．Zool．Listy9：159−166．