79 



made to confirm this idea, in order lo 

 assess tlie laxonomic position ol R. zhc/r^i 

 and further ascertain the invalidity of 

 Psetulorcinct, \sc tried in I lie present study 

 to clarify the phyloijenelic relationships of 

 R. zheiiv.! with icn species of brown frogs 

 from lapaii, laiwan, Russia, and Emopc 

 on I he basis scc|iiciice variation in the 

 niilochoneli lal l\ loclii onic b jjcnc. 



M \ 1 1 RI \I S AND Ml I HODS 



Wc extracted DNA I'roin small amounts 

 (less than 50 mg) of ethano! preserved tissue 

 samples from li\er and muscles of one R. 

 zhengi from Sichuan, C hina. Tissues from 

 specimens of A', wm/'orc/rni and /\. dalinci- 

 iiihi from { /cell and Austria ucic also used 

 foi ihc aiiab scs (sec ap|"icndi\). \\ c ampli- 

 IIclI a pail ot ihc mil)N\ cNloclimmc h 

 gene UMiiL' ilic poKmciasc chain rcaciioii 

 (PC'R) aiul sci.|ucnccd apjn (i\imalcl\ CiOO lip 

 of this L'cnc lollowiiiL' laiiaka ci al. ll^^M, 

 19^)6). I'rimcis used loi llic ampli fical ion 

 and sequencing were: 1.14850-1 (5'- 

 1 1 I C C GC \T(i..\T(;.\,\ AC TTC CiGCTC 

 -3': Tanaka ci al.. 1^)94), I 14X50-2 (5'- 

 TCTCA I ( ( 1 ( , \ I ( , \ \ \( I 1 1 (,( ,( re- 

 s': Tanaka cl al.. I'^'M. and I I 1 ^-^o: (5'- 

 G(i \ I i \( .( 1 ( i( I 1 ( I I ( I \ \ \ 1 I ( , 1 C 1 ( , 

 ( i( I 1 .m.ika I cno ci al., l^'JSb). W c 

 .idopicd ilic si.\|iicncc nimibcMii!.' s\sicm ol 

 iIk- limiian si,\|iiciicc (\iulcison ci al.. 

 \ ), 



We csiimaled genetic distance among 

 haplotypes based on the pairwisc maiii\ of 

 nucleotide sequence divergences calculated 

 bv Kimura's (I9S0) two-parameter meilunl 

 uiih the DNADIST program in IMHI II' 

 Version 3.5c ({•'elsenstein. 1993). .A distance 

 tree was generated with the neighbor-joining 

 algorithm (Saitou and Nei, 1987) in the 

 M K.lUioK program of ihe VWW IP 

 package 1 1 elseiistein, 199^). I oi maxunum 



likelihood (MI.) analysis, we employed 

 lelsenslein's (1993) DNAML algorithm 

 with empirical base frequencies and an 

 expected iransition-transversion bias of 

 2.0. Degrees of support for internal 

 branches in each tree were assessed by 1000 

 bootstrap pseudoreplicalions (Felsenstein, 

 1985). Maximum parsimony (MP) analysis 

 of unweighted, unordered character-stale 

 data was peiroinied \\ilh I he heuristic 

 search option using ilie iiee bisection- 

 recomieeiion (I BR) hraiieh-swapping in 

 PAUP* version 4. Ob (Swofford, 1998). 

 Confidence in each node was assessed by 

 1000 bootstraj-) i epliealions. 



For comparisons, we used the published 

 sequences of two populations of R. sauieri 

 (sensu Chou and I m, 1997) from Taiwan, 

 six species from Japan (R. pirica, R. ornal- 

 ivemris. R. juponicu. R. okinavana, R. 

 lii'^oi, and /\. i\ii\hiniciisis), one species 

 lioiii Russia (/\. unuirt'iisis), and an 

 oiiiL:roii|^ ia\on, R. caicsbeicina (Tanaka- 

 L eno el al., l99,Sa, b). We also incorpo- 

 rated a published sequence of Xenopti.s 

 laevis ( Duiion-Hluteau et al., 1985) into the 

 analysis as ihai lor another ontgroup. 



Rl si i TS 



We could constantly obtain nucleotide 

 sequence ilaia of 58^ bp for all samples 

 dig. 1). Samples of A'. (Uilnntlina from 

 ( /cell and .Austria showed an idenlical 

 sequence. In the neighbor-joining tree 

 (1 ig. 2.A), monophyly of R. zhengi and ihc 

 brown frogs was supported in high itera- 

 tions of 88. {"o. Nevertheless. R. zhfUfH 

 exhibited the earliest divergence among ihc 

 ingroup inxa and the monophyly of ihc 

 brown frogs was also supported in hii*h 

 iterations of S4.0<''o. 



Among ihc brown frogs, two I uio|K.m 

 species formed a cluster and iln •• • ! i . x' 



Fig. I. Aligned sequences of a 587 t>i 

 with ihe .sequence of Xenopus laevn 



1' I'! til',- 



