Detection of the apomictic mode of reproduction in maize-Tripsacum hybrids using maize RFLP markers

Polyploid plants in the genus Tripsacum, a wild relative of maize, reproduce through gametophytic apomixis of the diplosporous type, an asexual mode of reproduction through seed. Moving gene(s) responsible for the apomictic trait into crop plants would open new areas in plant breeding and agriculture. Efforts to transfer apomixis from Tripsacum into maize at CIMMYT resulted in numerou intergeneric F₁ hybrids obtained from various Tripsacum species. A bulk-segregant analysis was carried out to identify molecular markers linked to diplospory in T. dactyloides. This was possible because of numerous genome similarities among related species in the Andropogoneae. On the basis of maize RFLP probes, three restriction fragments co-segregating with diplospory were identified in one maize-Tripsacum dactyloides F₁ population that segregated 1∶1 for the mode of reproduction. The markers were also found to be linked in the maize RFLP map, on the distal end of the long arm of chromosome 6. These results support a simple inheritance of diplospory in Tripsacum. Manipulation of the mode of reproduction in maize-Tripsacum backcross generations, and implications for the transfer of apomixis into maize, are discussed.     

Heterochronic features of the female germline among several sexual diploid <Emphasis Type="Italic">Tripsacum</Emphasis> L. (Andropogoneae,Poaceae)

One element of gametophytic apomixis is unreduced embryo sac (ES) formation, which often occurs precociously displacing or replacing meiosis and causing apospory or diplospory, respectively. This study evaluated a premise that apomixis may evolve in hybridogenous plants that contain duplicate sets of allelically divergent ovule development heterochrony genes. The duplicate sets of genes would belong to duplicate genomic regions that are recombinationally isolated from each other (no gene flow) by allopolyploidy or paleopolyploidy, and this isolation would genetically stabilize apomixis. For apomixis to evolve, the ancestral donors of the duplicate regions must have differed from each other in timing of megasporogenesis, ES formation and embryony such that epigenetic misexpressions, or competitions in expression, of the duplicate heterochrony genes in hybridogenous derivatives would cause apomixis. Herein, we report substantial heterochrony in onset timing of germline stages among several sexual diploid Tripsacum genotypes, which may have been progenitors of apomictic polyploid Tripsacum. Tripsacum floridanum and Tripsacum zopilotense genotypes entered meiosis early. The former advanced rapidly through ES formation, but the latter entered a lengthy lag phase prior to ES formation. In two Tripsacum dactyloides var. dactyloides genotypes, meiosis occurred late and was followed by a distinct lag phase prior to ES formation. Likewise, the T. dactyloides var. meridonale genotype entered meiosis late, but the lag phase was brief. These differences appear to reflect allelic diversity at loci responsible for onset timing of different germline development stages within and across species and possibly across the recombinationally isolated duplicate chromosome regions in the Tripsacum paleopolyploid haplome (x = 18). Unique combinations of divergent alleles in hybridogenous plants coupled with polyploidy induced gene misexpressions may be required for apomixis to evolve. Electronic supplementary material Supplementary material is available in the online version of this article at and is accessible for authorized users.     

A comparison of apomictic reproduction in eastern gamagrass (Tripsacum dactyloides (L.) L.) and maize-Tripsacum hybrids

The expression of gene(s) governing apomictic reproduction inTripsacum provides the best foundation for comparing the effectiveness of apomictic reproduction in a series of maize-Tripsacum hybrids. Several 38-chromosome, apomictic maize-Tripsacum hybrids are available which possess the gene(s) conferring apomictic reproduction fromTripsacum. Without a base line for comparison, studies directed towards discerning the successful transfer or effectiveness of gene expression in a maize background are hampered. The objectives of this study are to compare the reproductive features found in apomicticTripsacum with those in apomictic maize-Tripsacum hybrids. In addition, this study determined the feasibility of utilizing these maize-Tripsacum hybrid materials to continue an attempt to transfer the genes into a pure maize background. The frequency and occurrence of five unique reproductive features found in apomictic accessions ofTripsacum dactyloides were compared to the reproductive behaviours exhibited in the maize-Tripsacum hybrids. Results indicate the genes controlling apomixis in tetraploidTripsacum are fully functional in maize-Tripsacum hybrids with diploid and triploid maize constitutions. The ability of theTripsacum apomictic genes to retain full expression provides evidence to continue their transfer to a diploid or tetraploid maize background.The use of company names in this publication does not imply endorsement by the USDA-ARS, or the product names or criticism of similar ones not mentioned. All programs and services of the U.S. Department of Agriculture are offered on a nondiscriminatory basis without regard to race, color, national origin, religion, sex, age, marital status, or handicap.     

Effect of pollen source and pollen ploidy on endosperm formation and seed set in pseudogamous apomictic Paspalum notatum

 It is generally accepted that most angiosperms require an accurate balance between maternal and paternal genome contribution for endosperm development. The endosperm balance number (EBN) hypothesis postulates that each species has an effective number which must be in a 2:1 maternal to paternal ratio for normal endosperm development and seed formation. The aim of this work was to investigate the effect of different sources and ploidy levels of pollen donors on endosperm formation and seed production of aposporous tetraploid (2n=4×=40) Paspalum notatum. Hand-emasculated spikelets of an apomictic 4× plant were dusted with pollen of 2×, 4×, 5×, 6× and 8× races of the same species; 3× and 4× races of a phylogenetically closely related species, P. cromyorrhizon; and 2× and 4× races of P. simplex, a species of a different subgenus. Experiments including self-pollination as well as emasculation without pollination were conducted for controls. Results indicated that apomictic 4×P. notatum is a pseudogamous species with effective fertilization of the two unreduced (2n) polar nuclei by a reduced (n) sperm. Endosperm development and seed production occurred independently of the species or the ploidy level of the pollen donor. However, seed germination rates were significantly lower than in the self-pollinated control when the pollen donor was 3×P. cromyorrhizon or 2× and 4×P. simplex. Aposporous embryo sacs in Paspalum contribute to endosperm formation with two unreduced (2n) polar nuclei, while the male contribution is the same as in sexual plants (n). Since sexual Paspalum plants fit the EBN hypothesis, the EBN insensitivity observed in apomictic plants might be a prerequisite for the spread of pseudogamous apomixis. The EBN insensitivity could have arisen as an imprinting consequence of a high maternal genome contribution. Received: 20 February 1998 / Revision accepted: 21 October 1998     

Apomixis via recombination of genome regions for apomeiosis (diplospory) and parthenogenesis in Erigeron (daisy fleabane, Asteraceae)

Apomixis in daisy fleabanes (Erigeron annuus and E. strigosus) is controlled by two genetically unlinked loci that regulate, independently, the formation of unreduced female gametophytes (apomeiosis, diplospory) and autonomous seed formation (parthenogenesis). In this work, fully apomictic F2s were regenerated by crossing F1s bearing, separately, these two functional regions. Two triploid (3x = 2n = 27) highly diplosporous F1s served as seed parents to an aneuploid (2x + 1 = 2n = 19) meiotic pollen donor bearing four AFLP markers linked to parthenogenetic seed formation but producing only abortive embryos and endosperm. Of 408 hybrids, 21 (5.1%) produced seed. Nine of these putative apomicts were tetraploids (4x), likely combining an unreduced egg from the diplosporous seed parent and a haploid gamete from the pollen parent (3x + x). The other 12 hybrid apomicts were pentaploid, interpreted as arising from the fusion of an unreduced diplosporous egg with an unreduced sperm cell (3x + 2x). Analysis indicated that all but three of the 21 synthetic apomicts recombined markers linked to diplospory and parthenogenesis. In addition, three additional hybrids combined markers linked to the two functional regions but produced only aborted embryos. The apomicts varied in percentage of diplosporous ovules (4.7–95.3% of all ovules produced) and in percentage of ovules that developed into seed (3.8–58.0%). These results support the hypothesis that apomeiosis and autonomous seed formation are genetically distinct, and that the traits can be separated and recombined to create hybrids exhibiting apomixis at near wildtype levels.     

Endosperm formation in aposporous Crataegus (Rosaceae, Spiraeoideae, tribe Pyreae): parallels to Ranunculaceae and Poaceae

Apomixis in Crataegus is primarily aposporous and requires pollination. The embryo sac is of the Polygonum type. A combination of meiotically unreduced embryo sacs with apparently reduced pollen would violate the usual requirement for a 2 : 1 ratio of maternal to paternal contributions to the endosperm. We therefore investigated the origin of endosperm in seeds of sexual diploids and apomictic polyploids of the sister genera Crataegus and Mespilus. Flow-cytometric DNA measurements from embryo and endosperm in mature seeds were converted to ploidy levels using leaf-tissue information. The diploids had triploid endosperm. In c. 60% of seed from polyploids, one sperm apparently contributes to the endosperm, while 25% or more may involve two sperm. Additional results suggest that trinucleate central cells also occur. Fertilization of meiotically unreduced eggs is indicated. The ratio of maternal to paternal contributions to the endosperm in these apomictic Crataegus is not constrained to 2 : 1. They thus resemble some Sorbus (Pyreae) and very distantly related Ranunculus (Ranunculaceae). It is suggested that Paspalum (Poaceae) may have similarly flexible endosperm ploidy levels.     

Apomixis is not developmentally conserved in related, genetically characterized Hieracium plants of varying ploidy

Apomixis is facultative in characterized members of the genus Hieracium. The three components that comprise the apomictic mechanism include apospory followed by autonomous embryo and endosperm formation. The time of aposporous embryo sac initiation and mode of embryo sac formation are different in Hieracium piloselloides (D3) and Hieracium aurantiacum (A3.4). Genetic studies have shown that a single dominant locus encodes all three components of apomixis in both species (Bicknell et al. 2000). We histologically examined a range of related, genetically characterized apomictic Hieracium plants derived from D3 and A3.4 to assess conservation of the apomictic mechanism in different genetic backgrounds. The plants varied in ploidy, and also in the amount of DNA introduced from sexual Hieracium pilosella (P4). An apomictic hybrid from a cross between the two apomicts was also examined. The developmental processes observed in the parental apomicts were not conserved in the examined plants and alterations occurred in the components of apomixis. One plant also exhibited adventitious embryony. The results show that other genetic factors can modify apomixis with respect to time of initiation, spatial location, and mode of developmental progression. Both the apomictic locus and the modifiers are essential for efficient penetrance of the trait in Hieracium. Some of the findings in Hieracium correspond with observations in Ranunculus and this is discussed in terms of models for apomictic development and the control of apomixis in crops. Received: 21 June 1999 / Revision accepted: 17 November 1999     

Tripsacum dactyloides (Poaceae): a natural model system to study parthenogenesis

Diplosporous apomeiosis, formation of unreduced embryo sacs primarily of the Antennaria type, followed by parthenogenetic embryo development and pseudogamy (fertilization of the central cell) describe gametophytic apomixis within the Tripsacum agamic complex. Tripsacum dactyloides (Eastern gamagrass) is a close relative of domesticated maize and was chosen as a natural model system to investigate gene expression patterns associated with parthenogenesis. The genome size of diploid sexual and polyploid apomictic T. dactyloides was estimated by flow cytometry to be 7.37 pg (2C), 14.74 pg (4C) and 22.39 pg (6C), respectively. The diploid genome size is thus approximately 1.352 larger than that of maize. The apomeiotic-pseudogamous pathway of seed formation was demonstrated at a rate of 92% by the flow cytometric seed screen (FCSS) with single mature seeds in tetraploid accessions. This number includes twin embryos which were detected in 13% of the seeds analyzed. Fertilization of unreduced egg cells (B_III hybrids) was measured in 10% of apomictic seeds. Autonomous (fertilization-independent) embryo development and fertilization-dependent endosperm formation were confirmed by pollination of tetraploid T. dactyloides with a diploid transgenic maize line carrying an actin::#-glucuronidase (GUS) reporter construct. GUS expression was detected after pollination in the developing endosperm, but not in the embryo. In similar intraspecific crossing experiments with maize, GUS expression was detected in both the embryo and endosperm. A protocol was established for microdissection of embryo sacs and early parthenogenetic embryos of T. dactyloides. Together, these techniques provide new tools for future studies aimed at comparing gene expression patterns between sexual maize and sexual or apomictic T. dactyloides.     

Non-Mendelian transmission of an apospory-specific genomic region in a reciprocal cross between sexual pearl millet (Pennisetum glaucum) and an apomictic F1 (P. glaucum×P. squamulatum)

We recently showed that aposporous apomixis, a form of gametophytic apomixis, is controlled by a single apospory-specific genomic region (ASGR) in both Pennisetum squamulatum and Cenchrus ciliaris. We present evidence that in a reciprocal cross between sexual pearl millet (P. glaucum) and an apomictic F1 (P. glaucum× P. squamulatum) the ASGR is not transmitted at the same rate. When pearl millet was used as the female parent and the apomictic genotype as the pollen donor, the ASGR was transmitted at a rate of 0.41 in a progeny of 57 plants, indicating a slight transmission ratio distortion. However, in a population of 52 rare sexual progenies characterized among a large progeny of a quasi-obligate apomict (an F1 hybrid of P. glaucum×P. squamulatum), the transmission rate of ASGR was only 0.12. This strong segregation distortion may have occurred at four different levels: (1) female meiosis, (2) during female gametophyte maturation, (3) upon fertilization with differential survival of embryos being a consequence of differential gene expression controlled by parent-of-origin specific effects (imprinting) and (4) at a later developmental stage of the embryo through an embryo/endosperm genetic incompatibility system. Recevied: 13 June 2000 / Revision accepted: 23 October, 2000     

The occurrence of phenotypically complementary apomixis-recombinants in crosses between sexual and apomictic dandelions (<Emphasis Type="Italic">Taraxacum officinale</Emphasis>)

Apomictic seed development in dandelion ( Taraxacum officinale) involves (1) restitutional meiosis (diplospory), (2) egg cell parthenogenesis, and (3) autonomous endosperm development. The question is whether these elements of apomixis are controlled by one single gene or by several independent genes. Five triploid non-apomictic hybrids, obtained in diploid sexual × triploid apomict crosses were characterized using cyto-embryological and genetic methods. Nomarski-differential interference contrast microscopy and the transmission of microsatellite markers and ploidy levels indicated that the hybrids combined elements of the apomictic and the sexual developmental pathway. Hybrids form two complementary groups with respect to the presence or absence of parthenogenesis and autonomous endosperm development. The occurrence of complementary apomixis-recombinants suggests that parthenogenesis and autonomous endosperm development in Taraxacum are regulated independently by different genes. This study also indicates that early embryo development is independent of endosperm formation, but that endosperm is essential for later embryo growth.     

Quantitative variation for apomictic reproduction in the genus Boechera (Brassicaceae)

? Premise of the study: The evolution of asexual seed production (apomixis) from sexual relatives is a great enigma of plant biology. The genus Boechera is ideal for studying apomixis because of its close relation to Arabidopsis and the occurrence of sexual and apomictic species at low ploidy levels (diploid and triploid). Apomixis is characterized by three components: unreduced embryo-sac formation (apomeiosis), fertilization-independent embryogenesis (parthenogenesis), and functional endosperm formation (pseudogamy or autonomous endosperm formation). Understanding the variation in these traits within and between species has been hindered by the laborious histological analyses required to analyze large numbers of samples. ? Methods: To quantify variability for the different components of apomictic seed development, we developed a high-throughput flow cytometric seed screen technique to measure embryo:endosperm ploidy in over 22000 single seeds derived from 71 accessions of diploid and triploid Boechera. ? Key results: Three interrelated features were identified within and among Boechera species: (1) variation for most traits associated with apomictic seed formation, (2) three levels of apomeiosis expression (low, high, obligate), and (3) correlations between apomeiosis and parthenogenesis/pseudogamy. ? Conclusions: The data presented here provide a framework for choosing specific genotypes for correlations with large "omics" data sets being collected for Boechera to study population structure, gene flow, and evolution of specific traits. We hypothesize that low levels of apomeiosis represent an ancestral condition of Boechera, whereas high apomeiosis levels may have been induced by global gene regulatory changes associated with hybridization.     

Evolution of gametophytic apomixis in flowering plants: an alternative model from Maloid Rosaceae

Gametophytic apomixis, asexual reproduction involving megagametophytes, occurs in many flowering-plant families and as several variant mechanisms. Developmental destabilization of sexual reproduction as a result of hybridization and/or polyploidy appears to be a general trigger for its evolution, but the evidence is complicated by ploidy-level changes and hybridization occurring with facultative apomixis. The repeated origins of polyploid apomictic complexes in the palaeopolyploid Maloid Rosaceae suggest a new model of evolutionary transitions that may have wider applicability. Two conjectures are fundamental to this model: (1) that as previously suggested by Rutishauser, like many sexual flowering plants the polyploid apomicts require maternal–paternal balance in the second fertilization event that gives rise to the endosperm, and (2) that the observed variation in endosperm ploidy levels relates less to flexibility late in development than to the known variation in developmental origin of the megagametophyte between mechanisms loosely categorized as diplospory and apospory. The model suggests explanations for the relative frequencies of apospory and diplospory, and for the wide but incomplete associations of apospory with a pollination requirement (pseudogamy) and of diplospory with autonomous development of the endosperm. It is suggested that pollination from other taxa may provide some adaptive advantage to pseudogamous apospory. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Heterochronic expression of sexual reproductive programs during apomictic development in Tripsacum

Some angiosperms reproduce by apomixis, a natural way of cloning through seeds. Apomictic plants bypass both meiosis and egg cell fertilization, producing progeny that are genetic replicas of the mother plant. In this report, we analyze reproductive development in Tripsacum dactyloides, an apomictic relative of maize, and in experimental apomictic hybrids between maize and Tripsacum. We show that apomictic reproduction is characterized by an alteration of developmental timing of both sporogenesis and early embryo development. The absence of female meiosis in apomictic Tripsacum results from an early termination of female meiosis. Similarly, parthenogenetic development of a maternal embryo in apomicts results from precocious induction of early embryogenesis events. We also show that male meiosis in apomicts is characterized by comparable asynchronous expression of developmental stages. Apomixis thus results in an array of possible phenotypes, including wild-type sexual development. Overall, our observations suggest that apomixis in Tripsacum is a heterochronic phenotype; i.e., it relies on a deregulation of the timing of reproductive events, rather than on the alteration of a specific component of the reproductive pathway.     

Mendelian segregation for two-factor apomixis in Erigeron annuus (Asteraceae)

The inheritance of asexual seed development (apomixis) in Erigeron annuus (Asteraceae) was evaluated in a triploid (2n=3x=27) population resulting from a cross between an apomictic tetraploid (2n=4x=36) pollen parent and a sexual diploid (2n=2x=18) seed parent. Diplospory (unreduced female gametophyte formation) and autonomous development (embryo and endosperm together) segregated independently in the population yielding four distinct phenotype classes: (1) apomictic plants combining diplospory and autonomous development, (2) diplosporous plants lacking autonomous development, (3) meiotic plants with autonomous (though abortive) development and (4) meiotic plants lacking autonomous development. Each class was represented by approximately one-quarter of the population (n=117), thus corresponding to a two-factor genetic model with no linkage (chi(2)=2.59, P=0.11). Observations demonstrate that autonomous embryo and endosperm development (jointly) may occur in either reduced or unreduced egg cells. The cosegregation of the traits is attributed to tight linkage or pleiotropy. The data are consistent with the hypothesis that autonomous development in E. annuus is regulated by a single fertilization factor, F, which initiates development of both the embryo and the endosperm in the absence of fertilization.     

Harnessing apomictic reproduction in grasses: what we have learned from Paspalum

Background

Apomixis is an alternative route of plant reproduction that produces individuals genetically identical to the mother plant through seeds. Apomixis is desirable in agriculture, because it guarantees the perpetuation of superior genotypes (i.e. heterotic hybrid seeds) by self-seeding without loss of hybrid vigour. The Paspalum genus, an archetypal model system for mining apomixis gene(s), is composed of about 370 species that have extremely diverse reproductive systems, including self-incompatibility, self-fertility, full sexual reproduction, and facultative or obligate apomixis. Barriers to interspecific hybridization are relaxed in this genus, allowing the production of new hybrids from many different parental combinations. Paspalum is also tolerant to various parental genome contributions to the endosperm, allowing analyses of how sexually reproducing crop species might escape from dosage effects in the endosperm.

Scope

In this article, the available literature characterizing apomixis in Paspalum spp. and its use in breeding is critically reviewed. In particular, a comparison is made across species of the structure and function of the genomic region controlling apomixis in order to identify a common core region shared by all apomictic Paspalum species and where apomixis genes are likely to be localized. Candidate genes are discussed, either as possible genetic determinants (including homologs to signal transduction and RNA methylation genes) or as downstream factors (such as cell-to-cell signalling and auxin response genes) depending, respectively, on their co-segregation with apomixis or less. Strategies to validate the role of candidate genes in apomictic process are also discussed, with special emphasis on plant transformation in natural apomictic species.     

A genetic linkage map of the diplosporous chromosomal region in<Emphasis Type="Italic"> Taraxacum officinale</Emphasis> (common dandelion; Asteraceae)

In this study, we mapped the diplosporous chromosomal region in Taraxacum officinale, by using amplified fragment length polymorphism technology (AFLP) in 73 plants from a segregating population. Taraxacum serves as a model system to investigate the genetics, ecology, and evolution of apomixis. The genus includes sexual diploid as well as apomictic polyploid, mostly triploid, plants. Apomictic Taraxacum is diplosporous, parthenogenetic, and has autonomous endosperm formation. Previous studies have indicated that these three apomixis elements are controlled by more than one locus in Taraxacum and that diplospory inherits as a dominant, monogenic trait (Ddd; DIP). A bulked segregant analysis provided 34 AFLP markers that were linked to DIP and were, together with two microsatellite markers, used for mapping the trait. The map length was 18.6 cM and markers were found on both sides of DIP, corresponding to 5.9 and 12.7 cM, respectively. None of the markers completely co-segregated with DIP. Eight markers were selected for PCR-based marker development, of which two were successfully converted. In contrast to all other mapping studies of apomeiosis to date, our results showed no evidence for suppression of recombination around the DIP locus in Taraxacum. No obvious evidence for sequence divergence between the DIP and non-DIP homologous loci was found, and no hemizygosity at the DIP locus was detected. These results may indicate that apomixis is relatively recent in Taraxacum.     

Cytogeography and reproduction of the Paspalum simplex polyploid complex

 Paspalum simplex is a grass distributed throughout the phytogeographic Chaco region in South America from which sexual diploid and apomictic tetraploid races have been reported. We analysed native populations to determine their homogeneity of ploidy level, and the relationship between geographic distribution, ploidy levels, and reproductive systems. The ploidy level was established for 379 plants from 32 wild populations. Tetraploidy and apomixis constitute the most common combination for this species all over the Chaco region. Apomictic hexaploid plants were found associated with 4x populations. Diploids were confined to a small sector of the region. One sexual triploid plant arose from seed harvested in a pure 2x population, and one apomictic 3x plant was found in a mixed 2x-4x population. The results suggest that P. simplex is a core agamic complex characteristic of the Chaco region from which other apomictic polyploid species of the subgenus Anachyris could have evolved. Received July 24, 2002; accepted September 12, 2002 Published online: December 11, 2002     

OBSERVATIONS ON INTROGRESSION OF TRIPSACUM INTO MAIZE

Derivatives of a cross between diploid Zea mays L. and Tripsacum dactyloides (L.) L. (2n = 72) were compared cytologically and morphologically. The objective of this study was to detect introgression from Tripsacum to maize that might have occurred during seven backcross generations with maize. Thirty-three morphological characters were used to analyze variation among aneuploid (20Zm + 2Td), 20-chromosome recovered maize, and the recurrent maize parent plants. Aneuploid and maize checks were extreme types, with 20-chromosome hybrid derivatives being morphologically intermediate. Several recovered maizes clustered with aneuploid plants and these hybrid derivatives have the greatest chance of Tripsacum introgression. Many traits such as endosperm abnormalities, tassel seed, albinos, tunicate glumes, tassel-tipped ears, fasciated and branched ear, and male spikelets between rows of kernels were observed. Although the genetic basis of many traits is unknown, mutations, epistatic effects or expression of Tripsacum chromatin are possible causes. The number of abnormal and tripsacoid traits observed in 20-chromosome recovered maizes indicates genetic transfer from Tripsacum to the maize genome.     

Is supernumerary chromatin involved in gametophytic apomixis of polyploid plants?

Gametophytic apomixis, or unreduced embryo sac development that results in asexual reproduction through seeds, occurs in several families of angiosperms and must be polyphyletic in origin. The molecular mechanisms underlying gametophytic apomixis have not been discovered and are the subject of intense investigation. A common feature of almost all apomicts is their polyploid nature. From genetic mapping studies in both monocots and dicots, there is low genetic recombination associated with a single (rarely two), dominant locus for either aposporous or diplosporous embryo sac formation. In Pennisetum squamulatum and Cenchrus ciliaris, some DNA sequences mapping to the apospory locus are unique to apomictic genotypes and apparently hemizygous. This sequence divergence at the apomixis locus could be a consequence of genome rearrangements and isolation from genetic recombination, both of which may have contributed to the definition of a chromosomal region as supernumerary. The possible involvement of supernumerary chromatin, formed as a result of interspecific hybridization, in the origin of apomixis, is explored here. Received: 26 October 2000 / Revision accepted: 5 April 2001