JOURNAL OF EXPERIMENTAL ZOOLOGY 283:295–306 (1999) Sex-Determining Mechanism in Buergeria buergeri (Anura, Rhacophoridae). III. Does the ZZW Triploid Frog Become Female or Male? S. OHTA,1* M. SUMIDA,2 AND M. NISHIOKA2 High School attached to Hiroshima-Denki Institute of Technology, Hiroshima 736-0022, Japan 2 Laboratory for Amphibian Biology, Faculty of Science, Hiroshima University, Higashihiroshima 739-8526, Japan 1 ABSTRACT Both triploids and gynogenetic diploids (GDs) were produced to clarify the relationship between the sex-chromosome constitution and the expression of sex in the common bellring frog, Buergeria buergeri. The sex differentiation of triploids in B. buergeri is quite remarkable. Triploid frogs consisted of three sex genotypes, ZZZ, ZWW and ZZW. All ZZZ triploids were males, and all ZWW triploids were females. It is very interesting that half of the ZZW triploids became female, and the other half became male. The GD frogs consisted of two sex genotypes, ZW and ZZ, which did not differ from the controls in sex differentiation. Since the ratios of ZZ and ZW eggs were significantly different among female parents, it is assumed that most (approximately 80– 90%) of the eggs made pre-reductional division in some females and post-reductional division in others during meiosis. It seems that ZW eggs were produced by the occurrence of recombination between the centromere and the sex-determining genes in B. buergeri. It was also found that the number of Z chromosomes in each cell of these triploids and GDs agreed with that of the nucleoli in each cell. J. Exp. Zool. 283:295306, 1999. © 1999 Wiley-Liss, Inc. Buergeria buergeri, which is endemic in Japan, is female heterozygous in sex determination, and chromosome pair no. 7 of this species is sex chromosomes of the ZZ/ZW-type (Ohta, ’82, ’86). The Z chromosome has a nucleolus organizer recognized as a secondary constriction and a satellite attached to the end of the long arm. On the other hand, the W chromosome does not have either the secondary constriction nor the satellite, making the distinction of the W chromosome from the Z chromosome easy. Schmid et al. (’93) showed that the nucleolus organizer region (NOR) and its associated constitutive heterochromatin are located exclusively in the long arm telomeric region of the Z chromosome, and not in the W chromosome or in any autosomes. Ohta (’87) obtained two mature sex-reversed genetic females from tadpoles administered with estrogen. Of the offspring from sex-reversed genetic females mated with normal females, 28% were ZZ zygotes, 49% were ZW zygotes, and the remaining 23% were WW zygotes which were mortal because they could not produce nucleoli. Recently, Atsumi et al. (’98) elucidated that the AAT-1 locus is sex-linked and that the codominant alleles are expressed on the Z chromosome, but not on the W chromosome in the Hiroshima population of B. buergeri. © 1999 WILEY-LISS, INC. Since triploid amphibians were first produced in an American newt Notophthalmus viridescens by Fankhauser and Griffiths (’39), numerous triploids have been obtained in many amphibian species by cold or heat-shock treatment of the eggs shortly after insemination in order to suppress extrusion of the second polar body (reviewed by Kawamura, ’84). The sex differentiation of artificially produced triploids has been also reported in several amphibian species. Fankhauser (’45) reported that most triploids produced by refrigeration in Notophthalmus viridescens were females. According to Humphrey and Fankhauser (’46), the triploids produced in Ambystoma mexicanum (axolotl) consisted of 19 females and seven males. When Ueda (’80) produced a large number of triploids in the toad Bombina orientalis, there was a nearly equal number of males and females among triploids. Triploids produced by cold or heat-shock treatment of fertilized eggs in Rana japonica and R. tsushimensis were all males (Kawamura and Tokunaga, ’52; Sumida and *Correspondence to: S. Ohta, High School attached to HiroshimaDenki Institute of Technology, Hiroshima 736-0022, Japan. Received 3 April 1998; Accepted 25 June 1998. 296 S. OHTA ET AL. Nishioka, ’93). A similar observation was made on triploid frogs of Rana rugosa from the Hiroshima population (Kashiwagi, ’93). Mature triploid frogs of Hyla japonica included an excessive number of males (Nishioka and Ueda, ’83). A nearly equal number of males and females was reported in triploids of Rana nigromaculata and R. brevipoda (Kawamura and Nishioka, ’67). Nevertheless, the relationship between the sex-chromosome constitution and the sex differentiation has never been reported in amphibian triploids except axolotl’s (Humphrey and Fankhauser, ’46). The present authors produced triploids of various sex-chromosome constitution using B. buergeri and examined their sexes in order to elucidate the relationship between the sex-chromosome constitution and the expression of sex in this species. Preliminary report of this research was made previously by Ohta (’88). MATERIALS AND METHODS Female and male B. buergeri (SCHLEGEL) were collected from Sandankyo, Togouchi, Hiroshima Prefecture. Triploids were produced principally according to the method of Nishioka (’72) by refrigerating fertilized eggs and suppressing extrusion of the second polar body. Refrigerating treatments were performed at 1–2°C for 3 hr, about 10 min after fertilization. Gynogenetic diploids were produced by refrigeration after insemination with ultraviolet irradiated sperm as the controls of the triploid production principally according to the method of Kawamura and Nishioka (’81). Toshiba GUL-5-J UV-lamp was used to inactivate the sperm nucleus. This lamp was operated at 68 V, 125 mA with a highest intensity of 254 nm. The effective time for inactivation of sperm nuclei exposed to UV-rays was 2 min. Somatic chromosomes were observed in the epidermal cells of the tail-tips of tadpoles by the techniques described previously (Ohta, ’86). The number of nucleoli of embryos or tadpoles was counted in each cell nucleus stained with Heidenhain’s iron hematoxylin or Giemsa solution (Ohta, ’86). Gonads were fixed in Navashin’s fluid, sectioned at 12 µm and stained with Heidenhain’s iron hematoxylin. RESULTS Production of triploids and gynogenetic diploids The mating experiments were carried out to produce triploids and GDs in the breeding season of the year of 1984. The developmental capacity of the offspring obtained from these matings was examined at six developmental stages from cleavage to completion of metamorphosis (Table 1). In order to determine the ploidy and sex-chromosome constitution of the offspring, karyotype and number of nucleoli were examined in each cell nucleus of the embryos or tadpoles (Tables 2 and 3). Controls Control matings were also made between nine females (84F, nos. 1–9) and nine males (84M, nos. 1–9) (Table 1). In these matings, 238 (56.9%) became normally feeding tadpoles. When the number of chromosomes was examined in 219 of these tadpoles, 218 were normal diploids having 26 chromosomes (Table 2). Of these diploid tadpoles, 115 (52.8%) were of ZW-type in sex-chromosome constitution and contained a single nucleolus in each cell (Table 3, Figs. 1b and 5a), while the other 103 (47.2%) were of ZZ-type and possessed a maximum of two nucleoli (Table 3, Figs. 1a and 5b). The remainder was a triploid having 39 chromosomes, TABLE 1. Developmental capacity of the eggs in two experimental series and the controls Parents Female 84F, nos. 1–9 84F, nos. 1–9 84F, nos. 1, 2, 4–7 1 Male 84M, nos. 1–9 84M, nos. 1–9 84M, nos. 1, 2, 4–7 Experimental series1 No. of eggs Control 418 Exp (3n) 1098 Exp (GD) 1451 No. of normally cleaved eggs (%) No. of normal neurulae (%) No. of normal tail-bud embryos (%) No. of normally hatched tadpoles (%) No. of normally feeding tadpoles (%) No. of metamorphosed frogs (%) 387 (92.6) 1014 (92.3) 1144 (78.8) 340 (81.3) 883 (80.4) 855 (58.9) 288 (68.9) 802 (73.0) 719 (49.6) 260 (62.2) 742 (67.6) 589 (40.6) 238 (56.9) 628 (57.2) 448 (30.9) 138 (33.0) 214 (19.5) 48 (3.3) Exp (3n), producing triploids by refrigeration after insemination with normal sperm; Exp (GD), producing gynogenetic diploids by refrigeration after insemination with UV-irradiated sperm. SEX OF ZZW TRIPLOIDS IN BELL-RING FROG 297 TABLE 2. Production of triploids and gynogenetic diploids Parents Female Male 84F, nos. 1–9 84F, nos. 1–9 84F, nos. 1, 2, 4–7 84M, nos. 1–9 84M, nos. 1–9 84M, nos. 1, 2, 4–7 Ploidy (%) Experimental series1 No. of tadpoles examined 2n 3n Mosaics Control Exp (3n) Exp (GD) 219 410 163 218 (99.5) 66 (16.1) 138 (84.7) 1 (0.5) 329 (80.2) — — 15 (3.7) 25 (15.3) 1 Exp (3n), producing triploids by refrigeration after insemination with normal sperm; Exp (GD), producing gynogenetic diploids by refrigeration after insemination with UV-irradiated sperm. and was ZZZ-type having a maximum of three nucleoli in each cell. It seems that this triploid was produced by extrusion failure of the second polar body nucleus from the egg. Triploids In order to produce triploids, a total of 1098 eggs obtained from nine females (84F, nos. 1–9) were refrigerated after insemination with the sperm of nine males (84M, nos. 1–9). There was no difference in the developmental capacity between the offspring from refrigerated eggs and the controls (Table 1). When the numbers of chromosomes and nucleoli in each cell were examined in 410 normally feeding tadpoles, 329 (80.2%) were triploids having 39 chromosomes, and 66 (16.1%) were diploids. The remaining 15 (3.7%) were diploid-triploid mosaics and diploid-tetraploid mosaics (Table 2). The examination of sex-chromosome constitution revealed that there were three sex genotypes, ZZW, ZWW and ZZZ, in these triploid tadpoles. Among the 329 triploid tadpoles, 84 (25.5%) were ZWW-type having a single nucleolus in each cell (Table 3, Figs. 2 and 5e), and 80 (24.3%) were ZZZtype having a maximum of three nucleoli (Table 3, Figs. 4 and 5g). The remaining 165 (50.2%) were ZZW-type having a maximum of two nucleoli in each cell nucleus (Table 3, Figs. 3 and 5f). ZWW and ZZZ zygotes predominated in the three, 84F, nos. 1–3 × 84 M, nos. 1–3, of the nine matings. Among the 170 triploids of these three matings, 78 (45.9%) were ZWW-type, 66 (38.3%) were ZZZ-type, and the remaining 26 (15.3%) were ZZW-type (Table 3). On the other hand, ZZW zygotes predominated in the other six matings, 84F, nos. 4–9 × 84M, nos. 4–9. Among the 159 triploids of these six matings, 139 (87.4%) were ZZW-type, six (3.8%) were ZWW-type, and 14 (8.8%) were ZZZ-type. Especially, among 73, 22 and 38 triploids obtained from three female parents (84F, nos. 5, 8 and 9), 62 (84.9%), 22 (100%) and 36 (94.7%) were ZZWtype, respectively (Table 3). Gynogenetic diploids Eggs of seven females (84F, nos. 1–7) were refrigerated after insemination with UV-irradiated sperm to produce GDs. None of the offspring raised from one female (84F, no. 3) developed beyond the neurula stage. Of 1451 eggs obtained from the re- TABLE 3. Sex chromosomes and nucleoli of triploids, gynogenetic diploids and the control tadpoles1 Control Parents Male Total ZW (1-nu) 84M, no. 1 84M, no. 2 84M, no. 3 84M, no. 4 84M, no. 5 84M, no. 6 84M, no. 7 84M, no. 8 84M, no. 9 32 23 5 4 59 7 20 40 28 21 12 2 1 26 4 12 22 15 Female 84F, no. 1 84F, no. 2 84F, no. 3 84F, no. 4 84F, no. 5 84F, no. 6 84F, no. 7 84F, no. 8 84F, no. 9 Total (%) 1 218 Triploids ZZ (2-nu) 11 11 3 3 33 3 8 18 13 115 103 (52.8) (47.2) The numbers in parentheses represent percentages. Gynogenetic diploids Total ZWW (1-nu) ZZZ (3-nu) ZZW (2-nu) Total ZW (1-nu) ZZ (2-nu) 69 69 32 12 73 6 8 22 38 31 30 17 0 3 2 1 0 0 24 33 9 0 8 0 4 0 2 14 6 6 12 62 4 3 22 36 1 12 — 38 70 7 10 — — 0 2 — 38 62 5 8 — — 1 10 — 0 8 2 2 — — 84 (25.5) 80 165 (24.3) (50.2) 329 138 115 (83.3) 23 (16.7) 298 S. OHTA ET AL. Fig. 1. Metaphase plates and karyotypes of a male and a female Buergeria buergeri. Scale bars equal 5 µm. a: Male. The pair of no. 7 chromosomes was homomorphic (ZZ-type). b: Female. The pair of no. 7 chromosomes was heteromorphic (ZW-type). maining matings between six females (84F, nos. 1, 2, 4–7) and six males (84M, nos. 1, 2, 4–7), 1144 (78.8%) cleaved normally (Table 1). At the neurula stage, 855 (58.9%) became normal embryos, but 289 (20.0%) were abnormal. Although 719 (49.6%) developed into normal tail-bud embryos, 136 (10.9%) were abnormal. The number of nucleoli in each cell nucleus was examined in the 289 abnormal neurulae and 136 abnormal tail-bud embryos. The results show that 124 neurulae and 94 tail-bud embryos, 218 in total, were WW zygotes having no nucleolus (Fig. 6c, d). WW individuals are mortal (Ohta, ’87). The remaining 207 neurulae and tailbud embryos had a single or a maximum of two nucleoli in each cell (Figs. 5c,d and 6a,b). Thereafter, 448 (30.9%) became normally feeding tadpoles (Table 1). When the chromosomes and nucleoli were examined in 163 normally feeding tadpoles, 138 were diploids, and the remaining 25 were diploidtetraploid mosaics and diploid-tetraploid-octoploid mosaics (Table 2). Among 138 diploids, 115 (83.3%) were ZW-type having a single nucleolus in each cell and the other 23 (16.7%) were ZZ-type having a maximum of two nucleoli in each cell in sex-chromosome constitution (Fig. 5c, d). Especially, among 12 diploids obtained from the female parent 84F, no. 2, ten (83.3%) were ZZ-type. On the other hand, among 38, 70 and 10 diploids obtained from three other female parents (84F, nos. 4, 5 and 7), 38 (100%), 62 (88.6%) and 8 (80.0%) were ZW-type, respectively (Table 3). Sex of triploids and gynogenetic diploids The gonads of frogs at ages of more than 3 months after metamorphosis were examined histologically (Table 4). The 17 ZW-type and 11 ZZtype GDs examined had similar ovaries and testes in terms of inner structure as those of the controls, respectively (Fig. 7a, b). These findings show that the sex differentiation in GDs is the same as that in the controls. Of triploids, all 33 ZWW-type frogs were females SEX OF ZZW TRIPLOIDS IN BELL-RING FROG 299 Fig. 2. Metaphase plate and karyotype of a ZWW-type triploid Buergeria buergeri. Scale bar equals 5 µm. The trip- let of no. 7 chromosomes consisted of one Z chromosome having a satellite and two W chromosomes having no satellite. having ovaries that were smaller than those of the controls at the same ages. Growing auxocytes were scarce and small, but there was little difference in the number of germ cells (Fig. 7c). All 53 ZZZ-type triploid frogs examined were males having testes. In these testes, the heads of the spermatozoa seemed to be somewhat long and thick compared with those of the controls (Fig. 7d). It was astonishing that there were both female and male ZZW zygotes, but no hermaphrodites were observed. Out of 80 ZZW-type triploid frogs examined, 43 (53.8%) were females having ovaries, and the remaining 37 (46.3%) were males having testes. This result shows that ZZW zygotes consisted of females and males at the rate of nearly 1:1 (Table 4). The ovaries of 43 ZZW females were small and underdeveloped compared with those of the controls, and were very similar 300 S. OHTA ET AL. Fig. 3. Metaphase plate and karyotype of a ZZW-type triploid Buergeria buergeri. Scale bar equals 5 µm. The triplet of no. 7 chromosomes consisted of two Z chromosomes having a satellite and one W chromosome having no satellite. to those of ZWW females in inner and outer structures (Fig. 7e). On the other hand, the testes of 37 ZZW males were somewhat larger than those of the controls at the corresponding age. Similar to the testes of ZZZ-type triploids, the heads of these spermatozoa were somewhat long and thick as compared with those of the controls (Fig. 7f). These findings show that triploids having two W chromosomes differentiate into females without fail, triploids having no W chromosomes differentiate into males, and triploids having a single W chromosome differentiate into either females or males. DISCUSSION Triploidy is established by a union of female and male pronuclei when the extrusion of the nucleus of the second polar body is suppressed by extremes of temperature. The sex differentiation of triploids has been reported in several amphibian species by many investigators. Fankhauser (’45) reported that most triploids produced by refrigeration in Notophthalmus viridescens were females. He explained this result with two hypotheses: the female is heterogametic, and post-reductional division for sex-determining segments of chromo- SEX OF ZZW TRIPLOIDS IN BELL-RING FROG 301 Fig. 4. Metaphase plate and karyotype of a ZZZ-type triploid Buergeria buergeri. Scale bar equals 5 µm. The triplet of no. 7 chromosomes consisted of three Z chromosomes having a satellite. somes occurred in most of the eggs, which were then expected to become ZZW-type female. Humphrey and Fankhauser (’46) produced triploids in Ambystoma mexicanum (axolotl) by using normal females (ZW), normal males (ZZ) and sex-reversed genetic females obtained by Humphrey (’42a,b, ’45). When the eggs of ZW females were refrigerated after insemination with sperm of ZZ males, the triploids consisted of 19 females and 7 males. These results have verified the correctness of the hypothesis that there are two types of reduction division, pre- and post-reductional divisions, for sex-determining segments of chromosomes during oogenesis. Cayrol et al. (’83) produced triploids from fertilized eggs by cold-shock treatment in Pleurodeles waltl, in which the female is heterogametic (Gallien, ’54; Collenot, ’73). They did not observe any significant modifications in the sex ratios of offspring of various crosses between the triploids. This result suggested pre-reduction at 302 S. OHTA ET AL. Fig. 5. Epidermal cells of triploid, GD and control tadpoles. Scale bars equal 5 µm. a: ZW-type control. b: ZZ-type control. c: ZW-type GD. d: ZZ-type GD. e: ZWW-type triploid. f: ZZW-type triploid. g: ZZZ-type triploid. meiosis leading to diploid eggs of the ZZ- or WWtype, and fertilization with Z sperm resulting ZZZ triploid females or ZWW triploid males. The present study clearly showed the relationship be- tween the sex-chromosome constitution and the expression of sex in triploid Buergeria buergeri. All 53 ZZZ triploids were males, and all 33 ZWW triploids were females. It is very interesting that Fig. 6. GD embryos at the tail-bud stage and their epidermal cells. Scale bars equal 2 mm (a, c) and 5 µm (b, d). a: ZZ- or ZW-type GD embryos. b: Epidermal cells of a ZZ-type GD embryo having a maximum of two nucleoli. c: WW-type GD embryos. d: Epidermal cells of a WW-type GD embryo having no nucleolus. The large black-spots stained vaguely with Giemsa solution were nuclei having no nucleolus. SEX OF ZZW TRIPLOIDS IN BELL-RING FROG 303 TABLE 4. Sex of triploids (3n), gynogenetic diploids (GD) and the control frogs Control Parents Gynogenetic diploids Male Total ZW X ZZ Y Total ZWW X ZZZ Y 84M, no. 1 84M, no. 2 84M, no. 3 84M, no. 4 84M, no. 5 84M, no. 6 84M, no. 7 84M, no. 8 84M, no. 9 31 22 4 — 9 — 6 36 18 19 11 2 — 4 — 2 20 8 12 11 2 — 5 — 4 16 10 41 42 9 4 14 6 8 10 32 13 15 2 0 1 1 1 0 0 17 24 4 0 4 0 4 0 0 3 1 3 4 8 4 3 7 10 126 66 60 166 33 53 43 Female 84F, no. 1 84F, no. 2 84F, no. 3 84F, no. 4 84F, no. 5 84F, no. 6 84F, no. 7 84F, no. 8 84F, no. 9 Triploids Total of 80 ZZW triploids, about half of them (43) became female, and the other half (37) became male. These results suggest that ZZZ and ZWW triploids were rigidly determined in sex differentiation, whereas ZZW triploids could differentiate into either males or females. The method for producing gynogenetic diploids has been applied to investigate various recessive mutations (Volpe and Dasgupta, ’62; Kawamura and Nishioka, ’77, ’81; Nishioka, ’77; Nishioka and Ueda, ’77, ’85a,b,c), and also to elucidate sex-determining mechanisms (Kawamura and Nishioka, ’77; Schmid et al., ’91). When Volpe and Dasgupta (’62) produced GDs by suppressing extrusion of the second polar body from eggs of burnsi mutants of Rana pipiens, 82.4% were burnsi mutants and only 17.6% were wild-type frogs. From this result, they assumed that recombination occurred with high frequency between the major burnsi locus and the centromere, and that the major burnsi gene made post-reductional division in most of the eggs. When Colombelli et al. (’84) obtained Xenopus laevis GDs by refrigerating eggs after insemination with UV-irradiated sperm, about 80% were females. To explain the predominance of females, they considered that many ZW zygotes were produced in addition to ZZ and WW zygotes in GDs due to the high frequency of recombination between the centromere and the sex-determining segments. The present authors obtained triploids by refrigerating eggs after fertilization with normal sperm and GDs by refrigerating eggs after insemination with UV-irradiated sperm in B. buergeri. As a result, ZZZ, ZWW and ZZW zygotes were obtained in triploids and ZZ, ZW and WW zygotes in GDs. These results suggest that the differential segments of the sex chromosomes make pre- or post-reductional division depending ZZW X Y Total ZW X ZZ Y 8 2 0 0 1 1 0 3 22 — 5 — 8 4 7 4 — — — 0 — 8 3 5 1 — — — 5 — 0 1 2 3 — — 37 28 17 11 on the eggs. Since the spermatozoa of the males of the present species have only Z chromosomes, it is clear that the ZZZ and the ZWW zygotes of triploids and the ZZ and the WW zygotes of GDs were produced from eggs in which sex chromosomes made pre-reductional division, and that the ZZW zygotes in triploids and the ZW zygotes in GDs were obtained from eggs in which sex chromosomes made post-reductional division. The distribution of chiasmata between the centromere and the nucleolus organizer was observed in the lampbrush bivalent chromosome no. VII in order to examine whether Z and W chromosomes segregate at pre- or post-reductional division in the oocytes of B. buergeri (Ohta, unpublished). These chiasmata could be interpreted as recombination sites between Z and W chromosomes. The nucleolus organizer was supposed to be closely linked with the sex-determining genes in this species (Ohta, ’86; Atsumi et al., ’98). The number of chiasmata was observed to be zero or one in the lampbrush bivalent no. VII by Ohta (unpublished). From this observation, the present authors assume that bivalent no. VII, which was converted into tetrad ZZWW, segregated into ZZ and WW in the case of no chiasmata between the centromere and the nucleolus organizer, whereas bivalent no. VII segregated into ZW and ZW in the case of one chiasma. It is supposed that ZZ and WW eggs were produced by pre-reductional division, whereas ZW eggs were produced by post-reductional division. In the present study, three kinds of triploids, ZZZ, ZWW and ZZW, were produced. Of these, ZZZ and ZWW triploids were considered to be derived from ZZ and WW eggs, respectively, having no chiasmata between the centromere and the nucleolus organizer in the bivalent no. VII. That is, ZZZ and ZWW triploids were produced by refrigerating these eggs and suppress- 304 S. OHTA ET AL. Fig. 7. Cross-sections of the gonads of triploid, GD and control Buergeria buergeri. Scale bars equal 50 µm. a: Ovary of a ZW-type GD frog. b: Testis of a ZZ-type GD frog. c: Ovary of a ZWW-type triploid frog. d: Testis of a ZZZ-type triploid frog. e: Ovary of a female ZZW-type triploid frog. f: Testis of a male ZZW-type triploid frog. ing extrusion of the second polar body after insemination with Z sperm. On the contrary, ZZW triploids were considered to be obtained by refrigerating ZW eggs, in which a recombination occurred between the centromere and the nucleolus organizer in the bivalent no. VII, inseminated with Z sperm. Moreover, it is certain that approximately 80– 90% of the eggs obtained from three females (84F, nos. 1, 2 and 3) made pre-reductional division, whereas the predominant majority of eggs (approximately 85–100%) laid by another four females (84F, nos. 4, 5, 8 and 9) made post-reductional division (Table 3). From these lines of evidence, it seems probable that there are great differences between SEX OF ZZW TRIPLOIDS IN BELL-RING FROG female parents in the rate of eggs made by prereductional and post-reductional division. Ohta (unpublished) examined the genotypes of triploid tadpoles obtained from one female frog (88F, no. 1) of the Hiroshima population and revealed that of 156 triploids, 127 (81.4%) were ZZZ or ZWW in sex genotype, and the other 29 (18.6%) were ZZW in sex genotype. When the bivalent no. VII was observed 4 months later in the number of chiasmata between the centromere and the nucleolus organizer, using the oocytes of the female frog (88 F, no. 1), the number of chiasmata was zero in 251 oocytes (88.1%) and one in 34 oocytes (11.9%). These results suggest that the number of chiasmata tends to inherit in this species. It is certain that the recombination was recognized as the chiasma, and that ZW eggs were produced by the occurrence of recombination between the centromere and the nucleolus organizer linked closely with the sex-determining genes in B. buergeri. Humphrey and Fankhauser (’46) conducted an interesting study on the relation between the sexchromosome constitution and the expression of sex in diploids and triploids of Ambystoma mexicanum (axolotl). According to them, triploids having only one W chromosome became female without exception. Namely, ZZW zygotes differentiated into females as did ZWW and WWW zygotes, but ZZZ zygotes differentiated into males. In the present study, all ZWW zygotes became female and all ZZZ zygotes became male. Although 80 ZZW triploid frogs were obtained, 43 of them were females and the remaining 37 were males. From these facts, it may be probable that female determining factors on the W chromosome of Buergeria buergeri are not always dominant for the Z chromosome, being different from axolotl’s. LITERATURE CITED Atsumi S, Ohta S, Sumida M. 1998. Sex differences in polymorphism and expression of AAT-1 in the Hiroshima population of Buergeria buergeri (Anura, Rhacophoridae). Zool Sci 15:477–483. Cayrol C, Jaylet A, Ferrier V, Gasser F. 1983. A genetic study of various enzyme polymorphisms in Pleurodeles waltlii (Urodele, Amphibian). III. The relationship between sexlinked peptidase-1 expression and gene-dose effects. Biochem Genet 21:551–559. Collenot A. 1973. Obtention, par la méthode des greffes de gonades embryonnaires, d’une femalle à descendance unisexuée femalle chez le Triton Pleurodeles waltlii Michah. Experientia 29:885–887. Colombelli B, Thiébaud ChH, Müller WP. 1984. Production of WW super females by diploid gynogenesis in Xenopus laevis. Mol Gen Genet 194:57–59. Fankhauser G. 1945. The effects of changes in chromosome 305 number on amphibian development. Quart Rev Biol 20: 20–78. Fankhauser G, Griffiths RB. 1939. Induction of triploidy and haploidy in the newt, Triturus viridescens, by cold treatment of unsegmented eggs. Proc Natl Acad Sci Wash 25: 233–238. Gallien L. 1954. Inversion expérimentale du sexe sous l’action des hormones sexuelles chez le Triton Pleurodeles waltlii Michah. Analyse des conséquences génétiques. Bull Biol France Belgique 88:1–51. Humphrey RR. 1942a. Sex of the offspring fathered by two Ambystoma females experimentally converted into males. Anat Rec 82(Suppl. 77):469. Humphrey RR. 1942b. Sex reversal and the genetics of sex determination in the axolotl (Ambystoma mexicanum). Anat Rec 84:465. Humphrey RR. 1945. Sex determination in ambystomid salamanders: a study of the progeny of females experimentally converted into males. Am J Anat 76:33–66. Humphrey RR, Fankhauser G. 1946. The development, structure and functional capacity of the ovaries in triploid ambystomid salamanders. J Morphol 79:467–510. Kashiwagi K. 1993. Production of triploids and their reproductive capacity in Rana rugosa. Sci Rep Lab Amphib Biol, Hiroshima Univ. 12:23–36. Kawamura T. 1984. Polyploid in amphibians. Zool Sci 1:1–15. Kawamura T, Nishioka M. 1967. On the sex and reproductive capacity of tetraploids in amphibians. Gunma Symp Endocrinol 4:23–39. Kawamura T, Nishioka M. 1977. Aspects of the reproductive biology of Japanese anurans. In: Taylor DH, Guttman SI, editors. The reproductive biology of amphibians. New York: Plenum Press. p 103–139. Kawamura T, Nishioka M. 1981. Developmental capacity and sex of gynogenetic diploids in Rana japonica and Rana tsushimensis. Sci Rep Lab Amphib Biol, Hiroshima Univ 5:325–358. Kawamura T, Tokunaga C. 1952. The sex of triploid frogs, Rana japonica Günther. J Sci, Hiroshima Univ Ser B Div 1 13:121–128. Nishioka M. 1972. The karyotypes of the two sibling species of Japanese pond frog, with special reference to those of the diploid and triploid hybrids. Sci Rep Lab Amphib Biol, Hiroshima Univ 1:319–337. Nishioka M. 1977. Color variants induced by radiation and their inheritance in Rana nigromaculata. Sci Rep Lab Amphib Biol, Hiroshima Univ 2:25–89. Nishioka M, Ueda H. 1977. Genetic and morphologic studies on ten albino stocks in Hyla arborea japonica. Sci Rep Lab Amphib Biol, Hiroshima Univ 2:103–163. Nishioka M, Ueda H. 1983. Studies on polyploidy in Japanese treefrogs. Sci Rep Lab Amphib Biol, Hiroshima Univ 6:207–252. Nishioka M, Ueda H. 1985a. Genetics and morphology of 13 albino stocks in the Rana nigromaculata group. Sci Rep Lab Amphib Biol, Hiroshima Univ 7:1–121. Nishioka M, Ueda H. 1985b. Two kinds of black-eyed variants in Hyla arborea japonica. Sci Rep Lab Amphib Biol, Hiroshima Univ 7:157–179. Nishioka M, Ueda H. 1985c. Non-inheritable color variants in Rana brevipoda porosa. Sci Rep Lab Amphib Biol, Hiroshima Univ 7:199–217. Ohta S. 1982. Heterozygosity of a lampbrush chromosome and sexual difference in the karyotype of Rhacophorus buergeri. Jpn J Genet 57:697–698 (In Japanese). 306 S. OHTA ET AL. Ohta S. 1986. Sex determining mechanism in Buergeria buergeri (SCHLEGEL). I. Heterozygosity of chromosome pair No. 7 in the female. Sci Rep Lab Amphib Biol, Hiroshima Univ 8:29–43. Ohta S. 1987. Sex determining mechanism in Buergeria buergeri (SCHLEGEL). II. The effects of sex hormones on the differentiation of gonads and offspring of sex-reversed females. Sci Rep Lab Amphib Biol, Hiroshima Univ 9:213–238. Ohta S. 1988. Triploids and gynogenetic diploids in Buergeria buergeri. Jpn J Genet 63:588 (in Japanese). Schmid M, Nanda I, Steinlein C, Kausch K, Epplen JT, Haaf T. 1991. Sex-determining mechanisms and sex chromosomes in amphibia. In: Green DM, Sessions SK, editors. Amphib- ian cytogenetics and evolution. San Diego: Academic Press. p 393–430. Schmid M, Ohta S, Steinlein C, Guttenbach M. 1993. Chromosome banding in Amphibia. XIX. Primitive ZW/ZZ sex chromosomes in Buergeria buergeri (Anura, Rhacophoridae). Cytogenet. Cell Genet 62:238–246. Sumida M, Nishioka M. 1993. Reproductive capacity of allotriploids between Rana tsushimensis from Tsushima and Rana japonica from Ichinoseki and Hiroshima. Sci Rep Lab Amphib Biol, Hiroshima Univ 12:133–175. Ueda H. 1980. The sex of triploids and gynogenetic diploids in Bombina orientalis. Sci Rep Lab Amphib Biol, Hiroshima Univ 4:185–199. Volpe EP, Dasgupta S. 1962. Gynogenetic diploids of mutant leopard frogs. J Exp Zool 151:287–302.