75 



Tahi r: 3. Measurements (in mm) and scale characters (if ilic Ryukyii live lined skink, Eumeces 

 i)Hiri!,iiuitus, from lolorishima Island. SVI,=snoiit-venl lenglh, IL tail leii.!Jili, i {. extent of the 

 Mici-liaii while line on tail, SR ninnhei ol inidbody scale rows, PN=postna.sal eondiiioii. 



Nn 



KUZ34568 

 KUZ34569 

 KUZ34570 

 KUZ34571 



*: I tail broken. 

 **: — not clear. 



I ).ik- Ml ( ollcLiion 



10 May 1997 

 10 May 1997 



10 May 1997 



11 May 1997 



Se\ S\ I 



lemale 4S.6 



male .'^1.5 



male 49.3 



teniale 48.8 



II 11 sk l'\ 



X9.4 -1/3 34 absent 



15.2 + * — ♦* 34 absent 



89.1 ^1/3 34 absent 



87.1 >l/3 34 absent 



(Tabic 3). These chaiacicr .states arc iden- 

 tical lo iht^sc in tlic two prc\imisly 

 coiiccicd specimens i cpoi lcd b\ i oyaina 

 (1995). 



Discussion 



Toyama (1995), in tiic description of 

 external characters in the two previous 

 specimens Irom lolorisliima KlaiKl, staled 

 liial then head sciitellalion seemingly fell 

 wiihm I lie rant-'c of variation in I 'uim'ccs 

 inur'j.uiiiiu\ . 11c Imihci noted lhal m these 

 specimens ihc cxicni ol ihc median wlnic 

 hnc on ilic lail, a chaiaclci used lo disciim- 

 malc ihc ivui Mihspecies of iiuir<^iiuiius , 

 was coiisisicni wilh lhal ol /• . o^huncnsis 

 lalhci llian liuil ol / . /n. nn/ri^ifuiliis . 

 ilouc\ci, he also noicd lhal ihc siiccimciis 

 h.id m,m\ more rows of midhod\ scales 

 (^4) ill, in / . imir<iimi(iis (24 ^(l). 1 \annna 

 lion ol ihc loui .uldiiional s|iccimcns 

 conlnmcd these morphological characteris- 

 tics loi the lolorisliima population of 

 Eumeces, and ihis alone ma\ suggesi a 

 genealogical distmcincss ol ihis popiil.u ion 

 from /•'. i/iiiri;irui!us . 



llo\\c\ci, icsuhs ol .maUses ol .illo/smc 

 d.u.i indicate th.ii ihc loionshima popiila 

 lion Is genetic. ill\ icm.iik.ibK similar to A. 

 marf:inaliis, ,iiul is nnuh closer to the 

 Tokimosliima |Hipiilation of /•.'. rn. oshi- 

 ///(■/(s/s than any other consiihspecitic popu- 

 laiuMis. This suggcst.s an evciiisiveh close 



hisiorica! affinily of ihc luo populaiions, 

 aiul \sc ihiis hclicxc lhal ihc loionshima 

 ixipnlalion should he classilicd as /. 

 osliinu'itsis , e\en though il cxhihiis a 

 uniL|uc slate in ihc miinbci of midbody 

 scale rows. Ii is probable ihai in such a 

 small island population as ihai on iotorish- 

 ima Island the midbody scale rows have 

 increased rapidly in response to selection 

 pressure from some cn\ ironmcnial factor 

 there. 



The KMorishnna |"io|nilalion has alleles 

 common lo ihc lokunoshima population 

 ai all loci and ihc 1) \aluc between these 

 po|iulaiions was \cr\ small (().()4.'5). Such 

 close L'cnciic icscmbl.ince between the 

 loioiishmia and lokunoshima populations 

 siiL'iJcsis lhal ilicsc populations luue been 

 \cr\ icccniU isolated. lloue\er, lotorish- 

 ima Island is suuated 65 kin wcsl of 

 lokunoshima Island, .md is separated 

 from ihe latier. as well as from other 

 Islands ha\ing mari;inaliis populations, 

 b\ straits no loss than 5(H) m in ilepih 

 (\ls \l. 19~S). Since sea level dropped by 

 1:0 140 in only during late Pleistocene 

 (Ota el al.. I99.>). ioiorishinia should have 

 icm.nned isolated then. The above result 

 thus suggests a relatively recent overseas 

 disjursal of the li/ard between these 

 islands. Otherwise, il necessitates an ad 

 hoc assumption for the formation of a land 

 bridge between them and its rcveni rapid 

 submergence. 



