Production of Fertile Unreduced Sperm by Hybrid Males of the Rutilus alburnoides Complex (Teleostei, Cyprinidae): An Alternative Route to Genome Tetraploidization in Unisexuals

Corresponding author: M. João Collares-Pereira, Centro de Biologia Ambiental, Departamento de Zoologia e Antropologia, Faculdade de Ciências, Universidade de Lisboa, Campo Grande C2 - Piso 3, 1700 Lisboa, Portugal., mcolares{at}fc.ul.pt (E-mail)

Communicating editor: S. YOKOYAMA

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The hybrid minnow Rutilus alburnoides comprises diploid and polyploid females and males. Previous studies revealed that diploid and triploid females exhibit altered oogenesis that does not involve random segregation and recombination of the genomes of the two ancestors, constituting unisexual lineages. In the present study, we investigated the reproductive mode of hybrid males from the Tejo basin, using experimental crosses and flow cytometric analysis of blood and sperm. The results suggest that diploid hybrids produced fertile unreduced sperm, transmitting their hybrid genome intact to offspring. Triploid hybrids also produced unreduced sperm, but it was not possible to obtain data concerning their fertility. Finally, tetraploid hybrids produced fertile diploid sperm, which exhibited Mendelian segregation. Tetraploid R. alburnoides may reestablish biparental reproduction, as individuals of both sexes with the appropriate constitution for normal meiosis (two haploid genomes from each parental species) are likely to occur in natural populations. Tetraploids probably have arisen from syngamy of diploid eggs and diploid sperm produced by diploid hybrid males. Diploid hybrid males may therefore play a significant role in the dynamics of the complex, starting the evolutionary process that may ultimately lead to a new sexually reproducing species.

INTERSPECIFIC hybridization generally leads either to sterile F₁'s or to hybrids with some measure of fertility that exhibit normal meiosis and act, through backcrossing, as a bridge for the transfer of genetic material between the parental species (introgression; reviewed in ARNOLD 1997 ). Occasionally, however, hybridization disrupts oogenesis so that hybrids produce viable eggs without recombination and often without a reduction in ploidy, founding unisexual lineages (reviewed in DAWLEY 1989 ). Among vertebrates, this phenomenon is apparently uncommon, with ~70 unisexual taxa described (VRIJENHOEK et al. 1989 ). The genetic basis of the origin of unisexuality via hybridization remains uncertain. It has been suggested that hybridization brings together specific sets of alleles that, in combination, cause changes in the regulation of oogenesis, allowing for unisexual reproduction ("balance hypothesis"; MORITZ et al. 1989 ). Genes seem to be differentially expressed in oocytes vs. spermatocytes, as rare males of unisexual lineages are apparently sterile (CIMINO and SCHULTZ 1970 ; DAREVSKY et al. 1978 ; RASCH et al. 1982 ; GODDARD and DAWLEY 1990 ). Rana esculenta is an exception, because both females and males are hybridogenetic (during gametogenesis, one of the parental genomes is eliminated without recombination, and only the other parental genome is retained in mature gametes; reviewed in GRAF and POLLS PELAZ 1989 ).

The Iberian minnow Rutilus (a.k.a. Tropidophoxinellus) alburnoides (Steindachner 1866) comprises diploid and polyploid forms and is of hybrid origin, incorporating genomes from Leuciscus carolitertii or L. pyrenaicus and that from an undescribed species (ALVES et al. 1997A , ALVES et al. 1997B ; CARMONA et al. 1997 ). Hybrid males are usually rare, but in some populations they represent about 30% of total specimens collected (COLLARES-PEREIRA 1984 , COLLARES-PEREIRA 1985 , COLLARES-PEREIRA 1989 ; ALVES et al. 1997A ; CARMONA et al. 1997 ). Previous studies have focused on the reproductive modes of diploid and triploid females, revealing altered oogenesis that does not involve random segregation and recombination between the genomes of the ancestors (CARMONA et al. 1997 ; ALVES et al. 1998 ). R. alburnoides females seem to exhibit distinct reproductive modes according to their geographic origin. Diploid and triploid females from the northern Douro basin reproduce by hybridogenesis, discarding the L. carolitertii genome during oogenesis (CARMONA et al. 1997 ). Diploid females from the southern Tejo basin transmit their hybrid genome clonally to the eggs, which upon fertilization yield triploid progeny, whereas triploid females from the Tejo and Guadiana basins present a modified hybridogenesis in which the L. pyrenaicus genome is discarded in each generation, but the remaining genomes are not transmitted clonally to the eggs as the elimination of the unmatched genome permits ready synapsis of the homospecific genomes ("meiotic hybridogenesis"; ALVES et al. 1998 ). Whether hybrid males are fertile and participate in the reproduction within the complex remained unknown.

As a key to exploring the population dynamics and the evolutionary history of the R. alburnoides complex, we have investigated the reproductive mode(s) of hybrid diploid, triploid, and tetraploid males from the Tejo basin. The present study is based on experimental crosses and flow cytometric analysis of blood and sperm and revealed a greater role for hybrid males than initially suspected.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Specimens used in this study were collected in 1996 from the Sorraia River of the Tejo basin (detailed locality data are available from M.J.C.P.), during the reproductive season (April–May).

DNA content of erythrocytes and spermatozoa from 31 R. alburnoides-like males was determined by flow cytometry, using an EPICS Profile II cytometer. Blood samples were drawn from the caudal vein, and milt samples were collected by applying light pressure to the abdomen. Both samples were stabilized in buffer (40 mM citric acid trisodium salt, 0.25 M sucrose, and 5% dimethyl sulfoxide) and immediately frozen at -80°. Prior to analysis, samples were diluted at approximately the same concentration (5 x 10⁶ cells/ml) with the help of a hemocytometer and stained as described in DAWLEY and GODDARD 1988 using propidium iodide as fluorochrome. Chicken erythrocytes were used both as an internal and external standard. The same sample of chicken erythrocytes was used throughout the study, and the flow cytometer was calibrated daily with fluorescent microspheres (DNA-Check, Coulter, Hialeah, FL). For each sample, two measurements were made at a flow rate of less than 300 events/sec, using an excitation emission of 488 nm. Approximately 6 x 10⁴ cells were examined per measurement, and only peaks of the fluorescence histograms with a coefficient of variation lower than 4% were scored (DRESSLER and SEAMER 1994 ). DNA content of erythrocytes and spermatozoa in individual fish was estimated by calculating the ratio of mean fluorescence of fish cells to mean fluorescence of chicken cells and multiplying by 2.5 pg (the standard DNA content value for chicken erythrocytes; TIERSCH et al. 1989 ). The data were converted and analyzed using the software Pro2FCS (version 3.2) and Win MDI (version 1.3.5).

Additional specimens were used in crossing experiments. Crosses were done blindly without knowledge of the ploidy level of the mates. For the crosses that involved hybrid males, progeny were produced by stripping eggs and milt from the parents simultaneously into a bowl containing aquarium water. The young hatched in <1 wk and were reared to the age of 9 mo (2–4 cm). High mortality occurred in all broods due to fungal contamination. Parents and offspring were killed with an overdose of MS222 and frozen at -80°. Offspring were sexed by dissection and inspection of gonads. Ploidy of parents and offspring was determined by flow cytometric measurement of erythrocyte DNA content as described above. Parents and a subsample of offspring were analyzed by DNA fingerprinting using the human minisatellite probes 33.6 and 33.15 (JEFFREYS et al. 1985 ), as described in ALVES et al. 1998 . The number of fragments detected by both probes out of n scored that were transmitted to a number r of offspring in a sibship of N was compared with the expected number given by the binomial distribution (^NCr/2^N)n assuming 50% transmission (JEFFREYS et al. 1986 ), using the G-test for goodness of fit (SOKAL and ROHLF 1981 ).

The hybrid nature of specimens analyzed in this study was investigated by allozyme electrophoresis at the sAAT* (aspartate aminotransferase, EC 2.6.1.1) and PGDH* (phosphogluconate dehydrogenase, EC 1.1.1.44) loci. These loci have been found to be diagnostic for the R. alburnoides complex; virtually all specimens are heterozygous, exhibiting one set of alleles associated with members of the genus Leuciscus and a second set that could not be attributed to any described species (ALVES et al. 1997A ; CARMONA et al. 1997 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Flow cytometric analysis of blood and sperm:
Flow cytometric measurement of DNA content in erythrocytes revealed that the 31 R. alburnoides-like males sampled in the Sorraia River comprised 21 diploids (mean DNA content of 2.43 ± 0.07 pg/cell), 2 triploids (mean DNA content of 3.64 ± 0.16 pg/cell), and 8 tetraploids (mean DNA content of 4.83 ± 0.08 pg/cell; Table 1). Ten diploid males produced sperm that showed DNA values ranging from 1.11 to 1.21 pg/cell, with a mean of 1.16 ± 0.03 pg/cell. The ratio of erythrocyte to spermatozoon DNA content per male varied from 1:0.44 to 1:0.50, averaging 1:0.48 ± 0.02. The remaining 11 diploid males produced sperm with DNA content ranging from 2.25 to 2.43 pg/cell, and the mean was 2.34 ± 0.06 pg/cell. Erythrocyte and spermatozoon DNA content ratio varied from 1:0.88 to 1:1, with a mean of 1:0.96 ± 0.03. Sperm produced by triploid males also exhibited DNA values similar to those of the erythrocytes (mean ratio 1:0.95 ± 0.03), averaging 3.44 ± 0.04 pg/cell. The DNA content of sperm produced by tetraploid males was from 2.19 to 2.57 pg/cell, with a mean of 2.34 ± 0.13 pg/cell. The ratio of the erythrocyte and spermatozoon DNA values ranged from 1:0.45 to 1:0.54, averaging 1: 0.48 ± 0.03.

Allozyme electrophoresis revealed that the diploid males with ratios of somatic cell and gamete DNA content of about 1:1 were all heterozygous at the diagnostic sAAT* and PGDH* loci, whereas the remaining diploids with ratios of about 1:0.5 were homozygous for alleles presumably attributed to the undescribed parental species (ALVES et al. 1997A ; CARMONA et al. 1997 ). All triploid and tetraploid males showed heterozygous patterns at the diagnostic loci.

The additional specimens used in the crossing experiments reported here included two diploid and four tetraploid males and one diploid and seven triploid females. All specimens were heterozygous at the diagnostic loci.

Crosses involving diploid hybrid males:
Diploid 121 crossed to diploid 122 yielded apparently all male tetraploid progeny (four fish were not sexed), whereas diploid 117 and 121 mated to triploid females produced triploid progeny of both sexes (Table 2).

DNA fingerprinting analysis of cross 122 x 121 (Figure 1a, Table 3) revealed that all offspring exhibited identical patterns, having coinherited all scorable maternal DNA fragments and almost all scorable paternal fragments. Probe 33.15 detected polymorphisms in the region around 9.4 kb, as one paternal fragment was transmitted to no offspring, and one fragment observed in four young could not be traced back to either parent.

View larger version (156K): In this window In a new window Download PPT slide   Figure 1. DNA fingerprints of parental fish and offspring of crosses 122 x 121 (a), and 122 x 130 (b), as obtained with probe 33.15 after MboI digestion. The arrowed fragments in offspring are present in neither parent and are probably new mutants; the arrowed fragment in female of cross #121 is present in no offspring and probably it was lost due to mutation.

View this table: In this window In a new window   Table 3. Segregation of hypervariable single fragments produced by the minisatellite probes 33.6 and 33.15 (JEFFREYS et al. 1985 ) in the experimental families

Crosses involving tetraploid hybrid males:
Tetraploid 130 and diploid 122 produced one brood of 24 tetraploid and 1 diploid young. Tetraploid offspring included both females and males; the single diploid was a female (Table 2). DNA fingerprinting analysis revealed that all tetraploid offspring exhibited bands associated with the male parent, but the diploid offspring displayed only bands that could be traced back to the mother (Figure 1b, Table 3). All scorable maternal DNA fragments were coinherited by all tetraploid and diploid progeny, indicating that the female hybrid genome was transmitted intact to the eggs. The observed distribution of the paternal bands in the tetraploid offspring followed the expected binomial distribution for alleles showing Mendelian inheritance (0.25 < P < 0.50). Probe 33.15 detected in four tetraploid progeny a fragment with length >9.4 kb that could not be attributed to either parent (Figure 1b).

Tetraploid 112, 115, and 134 crossed to triploid females resulted in only triploid progeny. Crosses 110 x 112, 111 x 112, and 133 x 134 produced offspring of both sexes, whereas cross 114 x 115 yielded apparently all female progeny (two fish were not sexed; Table 2). Families fathered by 115 and 134 were analyzed by DNA fingerprinting (Table 3). In each cross, sets of five and six maternal minisatellite bands, respectively, were transmitted to no offspring. The remaining maternal fragments and the paternal fragments were inherited by the triploid offspring following the binomial distribution for alleles segregating in Mendelian fashion (P > 0.1).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The measurement of DNA content in erythrocytes and spermatozoa demonstrated that some diploid R. alburnoides-like males produced haploid sperm (ratios of somatic cell and gamete DNA content of about 1:0.5), whereas others produced unreduced sperm (ratios of about 1:1). Allozyme analysis suggested that the former males possessed nonhybrid nuclear DNA, whereas the latter exhibited hybrid nuclear DNA. The origin of the nonhybrid diploid males lacks consensus among authors. Taking into account that the Hardy-Weinberg analysis of polymorphic loci revealed no significant deviations from random expectations, CARMONA et al. 1997 suggested that nonhybrid specimens correspond to a sexually reproducing diploid form, which most likely was the maternal ancestor of the complex. ALVES et al. 1998 (M. J. ALVES, M. J. COLLARES-PEREIRA, T. E. DOWLING and M. M. COELHO, unpublished data), considering that nonhybrid specimens exhibit L. pyrenaicus-like mitochondrial DNA and show highly male-biased sex ratios, suggested that they probably were reconstituted from R. alburnoides hybrids. Previous studies revealed that nonhybrid males exhibit normal Mendelian meiosis (ALVES et al. 1998 ), which is consistent with either of the above hypotheses.

DNA fingerprinting of progeny fathered by the diploid hybrid revealed that each inherited the almost intact genome of the male. These data together with the flow cytometry data clearly show that, like the diploid hybrid females from the Tejo basin (ALVES et al. 1998 ), diploid hybrid males produced unreduced clonal gametes. Although it has been hypothesized for Rana esculenta (GUNTHER 1970 ; UZZELL et al. 1977 ), production of fertile unreduced sperm by natural diploid hybrid vertebrates has never before been confirmed.

Triploid R. alburnoides males produced unreduced sperm (ratios of somatic cell and gamete DNA content of about 1:1). However, due to their low number in natural populations, they were not used in the crossing experiments, and no data concerning their fertility was obtained. There are reports of fertile triploid hybrid males in amphibians, but they all produced reduced sperm (HEPPICH et al. 1982 ; NISHIOKA and OHTANI 1984 ; SHEN et al. 1984 ; VINOGRADOV et al. 1990 ; BERGER and GUNTHER 1991 –1992; SUMIDA and NISHIOKA 1993 ).

Tetraploid R. alburnoides males produced reduced sperm (ratios of somatic cell and gamete DNA content of about 1:0.5) that exhibited Mendelian segregation at the minisatellite markers. Therefore, these males seem to have undergone normal meiosis. Theoretically, R. alburnoides tetraploids have three possible combinations of parental genomes: PPAA, PAAA, and PPPA, where P is the L. pyrenaicus genome and A is the genome of the other ancestor. Only in the first of these is normal meiosis likely to occur, as the presence of two haploid genomes from each parental species permits ready synapsis (VASIL'EV et al. 1989 ). We may, therefore, expect that the tetraploid males presently analyzed had a PPAA constitution.

The inheritance patterns of the maternal bands of both the diploid and triploid females used in the present study followed the patterns described in ALVES et al. 1998 . Diploid females from the Tejo basin transmit clonally their hybrid genome to the eggs, whereas triploid females present a modified hybridogenesis in which the L. pyrenaicus genome is discarded in each generation, but the remaining genomes undergo meiosis (meiotic hybridogenesis). According to ALVES et al. 1998 , eggs produced by both diploid and triploid females needed syngamy to initiate development. However, the present study revealed that a small proportion (3% in the present example) of the diploid eggs produced by diploid females may develop directly without syngamy into diploid clones. In fact, diploid 122 produced one diploid female young that inherited the intact hybrid genome of the female parent but exhibited no DNA fingerprinting bands that could be traced back to the father. Such young seems to have originated by gynogenesis, where sperm only stimulates development of the egg. This is the first direct evidence for gynogenesis in the R. alburnoides complex, constituting the third mode of reproduction described for diploid females (CARMONA et al. 1997 ; ALVES et al. 1998 ).

Probe 33.15 detected polymorphisms at crosses 122 x 121 and 122 x 130 that are probably accounted for by germ line mutation. Mutations in minisatellites can involve gain or loss of numbers of repeat units, originating new length alleles (JEFFREYS et al. 1988 ). In cross 122 x 121, one paternal fragment was transmitted to no offspring, while one fragment observed in some young could not be traced back to either parent. In cross 122 x 130, a new fragment was also observed in some offspring. According to BRUFORD et al. 1992 , typical apparent mutation rates in fingerprints are of the order of 10^-3 per fragment per gamete. Evidence for genomic variability in as short a time as one generation has also been found in one diploid R. alburnoides female (ALVES et al. 1998 ).

The mechanism of sex determination in the R. alburnoides complex is not well understood. Although cytological data indicated that the parental species L. pyrenaicus has a ZW female/ZZ male sex chromosome heteromorphism (COLLARES-PEREIRA et al. 1998 ), the sex ratio of offspring mothered by triploid R. alburnoides females (ALVES et al. 1998 ) has suggested that non-W-linked genes, expressed differently depending on the species and the population to which parents belong, are involved in sex determination. The present data are also not consistent with a simple ZW/ZZ sex-determining mechanism. According to this model, all progeny produced in cross 122 x 121 should have been female and not male, because they have received both a Z and W chromosome from their diploid mother. In fact, the occurrence of apparently all male progeny suggests an XX female/XY male sex chromosome heteromorphism. Male heterogamety would also explain the production of female and male offspring in the cross 122 x 130: because tetraploid males seem to undergo normal meiosis, we may expect that some spermatozoa carried two X chromosomes producing females, whereas others carried X and Y chromosomes producing males. However, this mechanism of sex determination does not explain the sex ratio of progeny of crosses 116 x 117 and 132 x 121, which involved diploid hybrid males and triploid females: according to this model, having received both an X and Y chromosome from their father, no female young should have been observed. Therefore, sex determination in R. alburnoides can be fully explained neither by a ZW female/ZZ male nor by an XX female/XY male sex-determining system.

SCHULTZ 1969 postulated that hybridization and unisexuality provide a path to evolution by polyploidy. Restoration of an even ploidy value in tetraploids would allow a return to normal meiosis, serving as a stepping-stone to biparental reproduction (SCHULTZ 1977 , SCHULTZ 1980 , SCHULTZ 1989 ). The typical route to tetraploidy in clonal vertebrates seems to be the syngamy of unreduced triploid eggs and haploid sperm produced by a male of a related bisexual species (DAWLEY 1989 ). This pathway could yield tetraploids with several possible combinations of parental genomes, though only one (two haploid genomes from each bisexual parental species) is optimal for meiotic reproduction (VASIL'EV et al. 1989 ). The formation of such tetraploids is, however, apparently not possible in many clonal vertebrates, as only one parental species is present (e.g., BOGART et al. 1987 ). In fact, although tetraploids are abundant in some unisexual complexes, namely in Ambystoma (BOGART 1989 ), Carassius auratus langsdorfii (KOBAYASI and KAWASIMA 1972 ) and Cobitis (VASIL'EV et al. 1989 ), all females seem to exhibit unisexual reproduction and the males, when they exist, are sterile. The R. alburnoides complex presents an alternative route to tetraploidy, where tetraploids may have originally arisen from syngamy of diploid eggs and diploid clonal sperm produced by a diploid hybrid male. At least two types of diploid eggs occur in the R. alburnoides complex, depending on their origin (ALVES et al. 1998 ): clonal eggs with PA constitution, if they are originated by diploid female hybrids; or recombined eggs with AA constitution, if they are produced by PAA triploid females. Fertilization of the former by clonal sperm yields isogenic tetraploids with PPAA constitution, whereas fertilization of the latter leads to genetically diverse tetraploids with PAAA constitution. The occurrence of matings between female and male diploid hybrids is restricted by the low frequency of these forms in natural populations, and we may speculate that PPAA tetraploids were initially infrequent. However, once they arose, they also produced fertile diploid gametes, yielding more tetraploids with the appropriate constitution for normal meiosis, either through backcrossing to diploid hybrids or through mating to other symmetric tetraploids. Symmetric tetraploids produced in both of these ways are genetically diverse, in contrast with clonal tetraploids whose parents are both diploid hybrids. Occurrence of PPAA tetraploids of both sexes in natural populations is the necessary condition for achieving biparental reproduction. The experimental hybridization between diploid hybrids reported here (cross 122 x 121, Table 2) produced apparently all male PPAA progeny. However, the cross involving a diploid female and a tetraploid male produced tetraploid females (cross 122 x 130) that should also present PPAA constitution, because they received a clonal PA genome from their mother and a recombined PA genome from their father. Therefore, we may expect that PPAA tetraploids of both sexes occur in nature, and that they may occasionally mate, leading to biparental reproduction.

Tetraploids seem to occur only in the central and northern part of the distribution area of the R. alburnoides complex, apparently being absent in the Guadiana and Sado basins (ALVES et al. 1997A , ALVES et al. 1997B ; CARMONA et al. 1997 ; MARTINS et al. 1998 ). One explanation is the apparent absence of diploid hybrid males in the latter basins (ALVES et al. 1997A ; CARMONA et al. 1997 ; M. J. ALVES, M. M. COELHO and M. J. COLLARES-PEREIRA, unpublished data). Therefore, in these basins, the only possible route to tetraploidy would involve the fertilization of triploid eggs. However, if such a phenomenon occurs, it seems to be very rare or localized.

The present study and the previous studies of CARMONA et al. 1997 and ALVES et al. 1998 suggest that hybridization in R. alburnoides has apparently "opened" many novel reproductive pathways, altering both oogenesis and spermatogenesis (see Figure 2 for summary of the putative reproductive modes of the different forms of the R. alburnoides complex in the Tejo basin). Contrasting with most unisexual vertebrates, hybrid R. alburnoides males are fertile and participate in the perpetuation of the Iberian complex. Among the known hybrid complexes, only in R. esculenta are hybrid males fertile (reviewed in GRAF and POLLS PELAZ 1989 ). Diploid hybrid males seem to have played a particularly important role in the dynamics and evolution of the R. alburnoides complex, as they might have initiated tetraploidization, starting the evolutionary process that may ultimately lead to a new sexually reproducing polyploid species.

View larger version (18K): In this window In a new window Download PPT slide   Figure 2. Summary of the putative reproductive modes and interrelationships between the different forms of the R. alburnoides complex in the Tejo basin (ALVES et al. 1998 ; present study). (—) Forms and pathways confirmed experimentally; (- - -) hypothetical forms and pathways. P is the genome of L. pyrenaicus, and A is the genome of the other ancestor. (1) Fertilization by P sperm, produced by normal meiosis by L. pyrenaicus; (2) fertilization by A sperm, produced by normal meiosis by nonhybrid males; and (3) fertilization by PA sperm, produced either clonally by diploid hybrid males or by normal meiosis by tetraploid males. (G) Occurrence of gynogenesis.

	ACKNOWLEDGMENTS

We are especially grateful to Toomas Saat for performing the breeding experiments. We thank Dominic Poccia for receiving M.J.C.P. and M.I.P. in his laboratory to improve technical aspects of flow cytometry, Terry Burke for receiving M.J.A. to learn DNA fingerprinting methods, and Diogo Thomaz and Olivier Hanotte for technical assistance during that time. We acknowledge Alec Jeffreys and Robert Dawley, who kindly provided the human minisatellite probes and the protocols for preservation and staining of erythrocyte samples for flow cytometry, respectively. We also thank Carlos Almaça and Eduardo Crespo for comments on the manuscript, Ricardo Pires and Luís Miguel Vieira for caring for the progeny on many occasions, Maria Graça Vieira for use of her laboratory facilities, and Líbia Zé-Zé and Maria do Céu Sampaio for technical assistance. We acknowledge Direcção Geral das Florestas for permission to collect specimens. This work was supported by Centro de Biologia Ambiental, by the Junta Nacional de Investigação Científica e Tecnológica (JNICT) project Programa Específico para o Ambiente (PEAM)/C/GAG/227/93, and by grants CIÊNCIA/BD/2185/92-RN and PRAXIS XXI/BD/5735/95 to M.J.A.

Manuscript received April 9, 1998; Accepted for publication September 18, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALVES, M. J., M. M. COELHO, and M. J. COLLARES-PEREIRA, 1997a  The Rutilus alburnoides complex (Cyprinidae): evidence for a hybrid origin. J. Zool. Syst. Evol. Res. 35:1-10.

ALVES, M. J., M. M. COELHO, M. J. COLLARES-PEREIRA, and T. E. DOWLING, 1997b  Maternal ancestry of the Rutilus alburnoides complex (Teleostei, Cyprinidae) as determined by analysis of cytochrome b sequences. Evolution 51:1584-1592.

ALVES, M. J., M. M. COELHO, and M. J. COLLARES-PEREIRA, 1998  Diversity in the reproductive modes of females of the Rutilus alburnoides complex (Teleostei, Cyprinidae): a way to avoid the genetic constraints of uniparentalism. Mol. Biol. Evol. 15:1233-1242.

ARNOLD, M. L., 1997 Natural Hybridization and Evolution. Oxford University Press, Oxford.

BERGER, L. and R. GÜNTHER, 1991  Inheritance patterns of water frog males from the environments of Nature Reserve Steckby, Germany. Zool. Pol. 37:87-100.

BOGART, J. P., 1989 A mechanism for interspecific gene exchange via all-female salamander hybrids, pp. 170–179 in Evolution and Ecology of Unisexual Vertebrates, edited by R. M. DAWLEY and J. P. BOGART. New York State Museum, Albany, NY.

BOGART, J. P., L. A. LOWCOCK, C. W. ZEYL, and B. K. MABLE, 1987  Genome constitution and reproductive biology of hybrid salamanders, genus Ambystoma, on Kelleys Island in Lake Erie. Can. J. Zool. 65:2188-2201.

BRUFORD, M. W., O. HANOTTE, J. F. Y. BROOKFIELD and T. BURKE, 1992 Single-locus and multilocus DNA fingerprinting, pp. 225–269 in Molecular Genetic Analysis of Populations: A Practical Approach, edited by A. R. HOELZEL. Oxford University Press, New York.

CARMONA, J. A., O. I. SANJUR, I. DOADRIO, A. MACHORDOM, and R. C. VRIJENHOEK, 1997  Hybridogenetic reproduction and maternal ancestry of polyploid Iberian fish: the Tropidophoxinellus alburnoides complex. Genetics 146:983-993[Abstract].

CIMINO, M. C., and R. J. SCHULTZ, 1970 Production of a diploid male offspring by a gynogenetic triploid fish of the genus Poeciliopsis. Copeia 1970: 760–763.

COLLARES-PEREIRA, M. J., 1984  The "Rutilus alburnoides (Steindachner, 1866) complex" (Pisces, Cyprinidae). I. Biometrical analysis of some Portuguese populations. Arq. Mus. Bocage Ser. A 2:111-143.

COLLARES-PEREIRA, M. J., 1985  The "Rutilus alburnoides (Steindachner, 1866) complex" (Pisces, Cyprinidae). II. First data on the karyology of a well-established diploid-triploid group. Arq. Mus. Bocage Ser. A 3:69-89.

COLLARES-PEREIRA, M. J., 1989 Hybridization in European cyprinids: evolutionary potential of unisexual populations, pp. 281–288 in Evolution and Ecology of Unisexual Vertebrates, edited by R. M. DAWLEY and J. P. BOGART. New York State Museum, Albany, NY.

COLLARES-PEREIRA, M. J., M. I. PRÓSPERO, R. I. BILÉU, and E. M. RODRIGUES, 1998  Leuciscus (Pisces, Cyprinidae) karyotypes: transect of Portuguese populations. Gen. Mol. Biol. 21:63-69.

DAREVSKY, I. S., L. A. KUPRIYANOVA and M. A. BAKRADZE, 1978 Occasional males and intersexes in parthenogenetic species of Caucasian rock lizards (genus Lacerta). Copeia 1978: 201–207.

DAWLEY, R. M., 1989 An introduction to unisexual vertebrates, pp. 1–18 in Evolution and Ecology of Unisexual Vertebrates, edited by R. M. DAWLEY and J. P. BOGART. New York State Museum, Albany, NY.

DAWLEY, R. M. and K. A. GODDARD, 1988  Diploid-triploid mosaics among unisexual hybrids of the minnows Phoxinus neogaeus.. Evolution 42:649-659.

DRESSLER, L. G., and L. C. SEAMER, 1994 Controls, standards, and histogram interpretation in DNA flow cytometry, pp. 241–262 in Flow Cytometry, edited by Z. DARZYNKIEWICZ, J. P. ROBINSON and H. A. CRISSMAN. Academic Press, San Diego.

GODDARD, K. A. and R. M. DAWLEY, 1990  Clonal inheritance of a diploid nuclear genome by a hybrid freshwater minnow (Phoxinus eos-neogaeus, Pisces: Cyprinidae). Evolution 44:1052-1065.

GRAF, J.-D., and M. POLLS PELAZ, 1989 Evolutionary genetics in the Rana esculenta complex, pp. 289–301 in Evolution and Ecology of Unisexual Vertebrates, edited by R. M. DAWLEY and J. P. BOGART. New York State Museum, Albany, NY.

GÜNTHER, R., 1970  Der Karyotyp von Rana ridibunda Pall. und das Vorkommen von Triploidie bei Rana esculenta L. (Anura, Amphibia). Biol. Zbl. 89:327-342.

HEPPICH, S., H. G. TUNNER, and J. GREILHUBER, 1982  Premeiotic chromosome doubling after genome elimination during spermatogenesis of the species hybrid Rana esculenta.. Theor. Appl. Genet. 61:101-104.

JEFFREYS, A. J., V. WILSON, and S. L. THEIN, 1985  Hypervariable "minisatellite" regions in human DNA. Nature 314:67-73[Medline].

JEFFREYS, A. J., V. WILSON, S. L. THEIN, D. J. WEATHERALL, and B. A. J. PONDER, 1986  DNA "Fingerprints" and segregation analysis of multiple markers in human pedigrees. Am. J. Hum. Genet. 39:11-24[Medline].

JEFFREYS, A. J., N. J. ROYLE, V. WILSON, and Z. WONG, 1988  Spontaneous mutation rates to new length alleles at tandem-repetitive hypervariable loci in human DNA. Nature 332:278-281[Medline].

KOBAYASI, H. and Y. KAWASIMA, 1972  On the chromosomes of an all-female population in the ginbuna, Carassius auratus langsdorfii.. Jpn. Wom. Univ. J. 17:259-263.

MARTINS, M. J., M. J. COLLARES-PEREIRA, I. G. COWX, and M. M. COELHO, 1998  Diploid vs. triploid R. alburnoides: spatial segregation and morphological differences. J. Fish Biol. 52:817-828.

MORITZ, C., W. M. BROWN, L. D. DENSMORE, J. W. WRIGHT, D. VYAS et al., 1989 Genetic diversity and the dynamics of hybrid parthenogenesis in Cnemidophorus (Teiidae) and Heteronotia (Gekkonidae), pp. 87–112 in Evolution and Ecology of Unisexual Vertebrates, edited by R. M. DAWLEY and J. P. BOGART. New York State Museum, Albany, NY.

NISHIOKA, M. and H. OHTANI, 1984  Hybridogenetic reproduction of allotriploids between Japanese and European pond frogs. Zool. Sci. 1:291-326.

RASCH, E. M., P. J. MONACO, and J. S. BALSANO, 1982  Cytophotometric and autoradiography evidence for functional apomixis in a gynogenetic fish, Poecilia formosa, and in its related triploid unisexuals. Histochemistry 73:515-533[Medline].

SCHULTZ, R. J., 1969  Hybridization, unisexuality and polyploidy in the teleost Poeciliopsis (Poeciliidae) and other vertebrates. Am. Nat. 103:605-619.

SCHULTZ, R. J., 1977  Evolution and the ecology of unisexual fishes. Evol. Biol. 10:277-331.

SCHULTZ, R. J., 1980 The role of polyploidy in the evolution of fishes, pp. 313–339 in Polyploidy: Biological Relevance, edited by W. H. LEWIS. Plenum Press, New York.

SCHULTZ, R. J., 1989 Origins and relationships of unisexuals poeciliids, pp. 69–87 in Ecology and Evolution of Livebearing Fishes (Poeciliidae), edited by G. F. MEFFE and F. F. SNELSON, JR. Prentice-Hall, Englewood Cliffs, NJ.

SHEN, J., Z. FAN, S. LI, R. CHENG, and S. XUE, 1984  Comparative studies of the somatic cell and spermatozoon DNA contents and ploidy of Fangzheng crucian carp and Zhalong lake goldfish. Acta Zool. Sin. 30:7-13.

SOKAL, R. R., and F. ROHLF, 1981 Biometry. Freeman, San Francisco.

SUMIDA, M. and M. NISHIOKA, 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.

TIERSCH, T. R., R. W. CHANDLER, S. S. WACHTEL, and S. ELIAS, 1989  Reference standards for flow cytometry and application in comparative studies of nuclear DNA content. Cytometry 10:706-710[Medline].

UZZELL, T., R. GÜNTHER, and L. BERGER, 1977  Rana ridibunda and Rana esculenta: a leaky hybridogenetic system (Amphibia Salientia). Proc. Acad. Nat. Sci. Phila. 12:147-171.

VASIL'EV, V. P., K. D. VASIL'EVA and A. G. OSINOV, 1989 Evolution of a diploid-triploid-tetraploid complex in fishes of the genus Cobitis (Pisces, Cobitidae), pp. 153–169 in Evolution and Ecology of Unisexual Vertebrates, edited by R. M. DAWLEY and J. P. BOGART. New York State Museum, Albany, NY.

VINOGRADOV, A. E., L. J. BORKIN, R. GÜNTHER, and J. M. ROSANOV, 1990  Genome elimination in diploid and triploid Rana esculenta males: cytological evidence from DNA flow cytometry. Genome 33:619-627[Medline].

VRIJENHOEK, R. C., R. M. DAWLEY, C. J. COLE and J. P. BOGART, 1989 A list of the known unisexual vertebrates, pp. 19–23 in Evolution and Ecology of Unisexual Vertebrates, edited by R. M. DAWLEY and J. P. BOGART. New York State Museum, Albany, NY.

This article has been cited by other articles:

				K. P. Lampert, D. K. Lamatsch, P. Fischer, and M. Schartl A Tetraploid Amazon Molly, Poecilia formosa J. Hered., March 1, 2008; 99(2): 223 - 226. [Abstract] [Full Text] [PDF]