127 



among R. I'lntcliikiriu , R. Jiincrcu, and A'. 

 arc'olulu, Inil conliadict ilic laiici in 

 showing ihc sole clivcrgLMicc ol A', nihidn. 

 Because llic results of Siies el al. (I9.S1) 

 were derived I'roin analyses ol' a relali\ely 

 small niunber ol loci for very small sam- 

 ples, we considei lliai our results more 

 successfullv rel'leet the ph\K\L'eii\ o\ Rliino- 

 ck'iiimys. 



The type species ol RluiH>clciiii)iys Fitz- 

 inger, 1835, is Tc\lu(li> dorsala Schoepff, 

 I SO I, which is a junior synonym ol A. 

 puiicniliiriii UeisDu, I'-^y^). I heret'ore, 

 pitiicliilaiiu and its lelatixes (urcohild , 

 fiirwrca, and /'iiUhcr/nnn) should be 

 retained luider Riiinoclciiiiitv'^, while itihuhi 

 [wpc species ol ( hclopiis) and unniilaiu 

 slunild he ino\ed lo the resurrected genus 

 C/ic/o/'us ( ope, IS""!), which was roinierU 

 regarded as a unnor s\non\ni ol R/ii/io- 

 ck'iiiniys (see Weimulh and Mertens, 

 1977). 



I'liri/IU'l cvoliiiion ()/ /i/asin/l /u/i'^e 



Wiihni the Hal apui idae, a kinetic phisiron 

 equipped with a lunge between the i")ectoral 

 and abdominal was obser\cd in five 

 gCtK'uudme genera, ( 'tsiDclciiiinvs , Cuoru, 

 CwU'iiivs, \()!<)chclys, and Pyxidca. Bram- 

 ble ( Ur4) tlemonstrated tlisiinet dilTerences 

 m shell elosuii' meeh,imsm between em\did 

 box tuitles ,uid b.uagurid box tunics, lie 

 argued th,ii in the ('ieoem\ tlinae the 

 plaslral hinge hael e\ol\cd two times, m ihe 

 common ancestor ot i istoclcinmys, Ciiora, 

 ( 'i(7(7/;v s, and Pvytdai. and in Solochclvs. 

 In .uldiiion. he noted ih.ii //idsruivs had a 

 moie developed muselus lestiiseaiiularis 

 l.i musele elemeni .issiK'iated with the 

 closuie ol the anieiioi |i|,isti.d lobe in the 

 bataguiid box tmiles: Bramble. 19-4) than 

 lllc other geoenu dines lacking a plasiial 

 hinge. He thus suspecteil llie dcii\alion 

 of 0,v/rn7( //;//n s, C 'tiorn, Cvch'tnys, aiul 

 Pyxiiifu trom a conunon //('(Mfrnvv-likc 

 anccsioi , .md, on the injsis of this consider- 

 ation, united the live genera in the Hcnsc- 

 mys group. However, oin stiuly showed 



no distinct dilTerences in the degree of 

 dc\elopmeni ol m. tesloseapularis between 

 Ucoscniys and other "non-box" geoemyd- 

 ines (character 21). Additionally, we did 

 not find an\ character to support mono- 

 phyly ol the I Icoscmys group, or ol an 

 assemblage ol' Cislovleniiuys, Cuorci, 

 Cyck'inys, and Pyxidca. We thus consider 

 Bramble's (1974) grouping as unnatural. 



Bramble (1974) also suggested that plas- 

 tral kinesis of Nuloclwlys arose apart from 

 the lU'usc'inys group. But he hardly 

 mentioned the difference in shell-closing 

 mechanism between Noiochclys and the 

 other batagurid box turtles, in adult living 

 specimens, both anterior and posterior 

 plastral lobes of Notochelys are as movable 

 as those of Cycleiuys but less movable 

 than those of CisioclcniiDvs , C'liorn. and 

 Pyxidca (Yasukawa, unpublished data). 

 Both Cyc/eniys and Woloc/wlys have no 

 hinge in juveniles, while the other three 

 genera have a weak hinge even in juveniles 

 (^■asukavva, unpublished data). In contrast, 

 the verv long antcrodorsal jiortion of the 

 iliac blade, a condition unique to the 

 bataguiids (ehaiacter state .^4 [1]: Fig. Ic, 

 d), was observed onlv in Cistoclemmys. 

 i he shape of the iliac blade appears to be 

 assoeiated with the cU>surc o\ tiie posterior 

 |Masii,il lobe iBiamble, 19-4). We thus 

 susjiecl that the most prominent difference 

 in shell-closing mechanism exists between 

 Cistoclemmys ami the other gcoeniydine 

 box turtles, not between S'onnfirlys and 

 the others. In the Cieoenndinae. the plas- 

 tial hinge seems to have acliially evolved 

 three times— in the .Wilochelys-Cyclcmys 

 clade (.Stem Ci), Cuoni (.Stem M). and ihc 

 Pyxidea-Cisfodemmys clade (Sicm Q). 



In adult females of sonic balagurids 

 lacking the plasiral hinge, llic bony connec- 

 tion bciwecn ihc hypoplasiron and carapace 

 is replaced by a liganjcnioiis connection. 

 .Such plaslral kinesis. sonielinu'N called 

 sexually dimorphic plaslral kinesis, is prob- 

 ably an adaptation for ihc passage of rela- 

 tively large eggs which olherwisc could noC 



