Sequence organisation analysis of the wheat and rye genomes by interspecies DNA/DNA hybridisation

Rapid and cost-effective molecular karyotyping in wheat, barley, and their cross-progeny by chromosome-specific multiplex PCR

BACKGROUND

Interspecific hybridisation is a powerful tool for increasing genetic diversity in plant breeding programmes. Hexaploid wheat (Triticum aestivum, 2n = 42) × barley (Hordeum vulgare, 2n = 14) intergeneric hybrids can contribute to the transfer of agronomically useful traits by creating chromosome addition or translocation lines as well as full hybrids. Information on the karyotype of hybrid progenies possessing various combinations of wheat and barley chromosomes is thus essential for the subsequent breeding steps. Since the standard technique of chromosome in situ hybridisation is labour-intensive and requires specific skills. a routine, cost-efficient, and technically less demanding approach is beneficial both for research and breeding.

RESULTS

We developed a Multiplex Polymerase Chain Reaction (MPCR) method to identify individual wheat and barley chromosomes. Chromosome-specific primer pairs were designed based on the whole genome sequences of 'Chinese Spring' wheat and 'Golden Promise' barley as reference cultivars. A pool of potential primers was generated by applying a 20-nucleotide sliding window with consecutive one-nucleotide shifts on the reference genomes. After filtering for optimal primer properties and defined amplicon sizes to produce an ordered ladder-like pattern, the primer pool was manually curated and sorted into four MPCR primer sets for the wheat A, B, and D sub-genomes, and for the barley genome. The designed MPCR primer sets showed high chromosome specificity in silico for the genome sequences of all 18 wheat and barley cultivars tested. The MPCR primers proved experimentally also chromosome-specific for the reference cultivars as well as for 13 additional wheat and four barley genotypes. Analyses of 16 wheat × barley F1 hybrid plants demonstrated that the MPCR primer sets enable the fast and one-step detection of all wheat and barley chromosomes. Finally, the established genotyping system was fully corroborated with the standard genomic in situ hybridisation (GISH) technique.

CONCLUSIONS

Wheat and barley chromosome-specific MPCR offers a fast, labour-friendly, and versatile alternative to molecular cytogenetic detection of individual chromosomes. This method is also suitable for the high-throughput analysis of distinct (sub)genomes, and, in contrast to GISH, can be performed with any tissue type. The designed primer sets proved to be highly chromosome-specific over a wide range of wheat and barley genotypes as well as in wheat × barley hybrids. The described primer design strategy can be extended to many species with precise genome sequence information.

Perspective: 50 years of plant chromosome biology

The past 50 years has been the greatest era of plant science discovery, and most of the discoveries have emerged from or been facilitated by our knowledge of plant chromosomes. At last we have descriptive and mechanistic outlines of the information in chromosomes that programs plant life. We had almost no such information 50 years ago when few had isolated DNA from any plant species. The important features of genes have been revealed through whole genome comparative genomics and testing of variants using transgenesis. Progress has been enabled by the development of technologies that had to be invented and then become widely available. Arabidopsis (Arabidopsis thaliana) and rice (Oryza sativa) have played extraordinary roles as model species. Unexpected evolutionary dramas were uncovered when learning that chromosomes have to manage constantly the vast numbers of potentially mutagenic families of transposons and other repeated sequences. The chromatin-based transcriptional and epigenetic mechanisms that co-evolved to manage the evolutionary drama as well as gene expression and 3-D nuclear architecture have been elucidated these past 20 years. This perspective traces some of the major developments with which I have become particularly familiar while seeking ways to improve crop plants. I draw some conclusions from this look-back over 50 years during which the scientific community has (i) exposed how chromosomes guard, readout, control, recombine, and transmit information that programs plant species, large and small, weed and crop, and (ii) modified the information in chromosomes for the purposes of genetic, physiological, and developmental analyses and plant improvement.

Rye chromosome-specific polymerase chain reaction products developed by primers designed from the EcoO109I recognition site

From our analysis of repeat sequences in the rye genome, the presence of multiple restriction sites of EcoO109I (5'-PuGGNCCPy-3') across the genome has been predicted. By first using primers designed to contain EcoO109I sites in polymerase chain reaction (PCR), polymorphic DNA markers were effectively obtained. A total of 43 types of 10-mer primers containing EcoO109I sites were applied for PCR by using genomic DNA of Secale cereale self-fertile line IR27 and Triticum aestivum 'Chinese Spring' (CS) as the template. Twenty two primers detected polymorphisms between wheat and rye, and they were applied for PCR using a series of CS wheat--'Imperial' rye chromosome addition lines as templates. Nine chromosome-specific amplification fragments identified on five chromosomes were collected from gels and hybridized with nylon membrane-transferred PCR products from the wheat-rye chromosome addition lines. The gel blot was only observed between the collected fragments; therefore, these fragments were confirmed to be chromosome-specific. These fragments were sequenced and converted to sequence-tagged site (STS) primers. We therefore introduce a new method for building chromosome-specific DNA markers: (i) multiple polymorphic fragments can be obtained from EcoO109I primers and (ii) the addition of three nucleotides to the EcoO109I site restricts the amplification region to generate chromosome-specific fragments.

Effective isolation of retrotransposons and repetitive DNA families from the wheat genome

New classes of repetitive DNA elements were effectively identified by isolating small fragments of the elements from the wheat genome. A wheat A genome library was constructed from Triticum monococcum by degenerate cleavage with EcoO109I, the recognition sites of which consisted of 5'-PuGGNCCPy-3' multi-sequences. Three novel repetitive sequences pTm6, pTm69 and pTm58 derived from the A genome were screened and tested for high copy number using a blotting approach. pTm6 showed identity with integrase domains of the barley Ty1-Copia-retrotransposon BARE-1 and pTm58 showed similarity to the barley Ty3-gypsy-like retrotransposon Romani. pTm69, however, constituted a tandem array with useful genomic specificities, but did not share any identity with known repetitive elements. This study also sought to isolate wheat D-genome-specific repetitive elements regardless of the level of methylation, by genomic subtraction. Total genomic DNA of Aegilops tauschii was cleaved into short fragments with a methylation-insensitive 4 bp cutter, MboI, and then common DNA sequences between Ae. tauschii and Triticum turgidum were subtracted by annealing with excess T. turgidum genomic DNA. The D genome repetitive sequence pAt1 was isolated and used to identify an additional novel repetitive sequence family from wheat bacterial artificial chromosomes with a size range of 1 395-1 850 bp. The methods successfully led pathfinding of two unique repetitive families.

Genus-specific localization of the TaiI family of tandem-repetitive sequences in either the centromeric or subtelomeric regions in Triticeae species (Poaceae) and its evolution in wheat

The TaiI family sequences are classified as tandem repetitive DNA sequences present in the genome of tribe Triticeae, and are localized in the centromeric regions of common wheat, but in the subtelomeric heterochromatic regions of Leymus racemosus and related species. In this study, we investigated the chromosomal distribution of TaiI family sequences in other Triticeae species. The results demonstrated a centromeric localization in genera Triticum and Aegilops and subtelomeric localization in other genera, thus showing a genus-dependent localization of TaiI family sequences in one or the other region. The copy numbers of TaiI family sequences in species in the same genus varied greatly, whether in the centromeric or subtelomeric regions (depending on genus). We also examined the evolution of TaiI family sequences during polyploidization of hexaploid common wheat. A comparison of chromosomal locations of the major TaiI family signals in common wheat and in its ancestral species suggested that the centromeric TaiI family sequences in common wheat were inherited from its ancestors with little modification, whereas a mixed origin for the B genome of common wheat was indicated.

Characterization of a highly repeated DNA component of perennial oats (Helictotrichon, Poaceae) with sequence similarity to a A-genome-specific satellite DNA of rice (Oryza)

The taxonomic relationships among perennial oats (Helictotrichon Besser ex Schultes & Schultes, Aveninae, Aveneae, Poaceae) have been studied using highly repeated satellite DNA as a molecular marker. Highly repetitive sequences were isolated from restriction endonuclease digests of nuclear DNA of Helictotrichon convolutum, and satellite repeats (approximately 365 bp in length) were cloned, sequenced and compared among each other. They exhibited an intraspecific sequence variability of 6-9%. This satellite DNA, CON1, is differentially distributed within the genus Helictotrichon. In species of the subgenus Helictotrichon a high copy number is detectable, whereas in representatives of the subgenera Pratavenastrum and Pubavenastrum the number of copies per genome is rather low. Surprisingly, the satellite DNA repeat CON1 shows 74% sequence similarity to an A-genome specific repetitive DNA of Oryza (rice).

Targeting deletion (homoeologous chromosome pairing locus) or addition line single copy sequences from cereal genomes

We describe here a protocol for obtaining clones containing sequences present in low copy-number from genomic DNA where moderately and highly repeated sequences predominate. Specific chromosomal regions can be targeted by using deletion or addition line material. We have used this protocol to identify a sequence which has been deleted in both the tetraploid and hexaploid wheat mutants for the homoeologous chromosome pairing locus.

BIS 1, a major component of the cereal genome and a tool for studying genomic organization

We report the cloning and characterization of an element (BIS 1) in barley. Related sequences were also found in wheat and rye genomes. BIS 1-related sequences may be some of the more frequent of the complex repeats in the barley genome, and their dispersion throughout the barley genome suggests that they are capable of both mobility and amplification. BIS 1 sequences have been used to study the gross structure of the barley genome. These studies indicate that the genome may be formed from larger genomic structures of tens of kilobases which may be repeated. Surprisingly, these studies also show that these large genomic structures also occur in similar proportions and at similar size distribution in the wheat genome.

In situ hybridization as a rapid means to assess meiotic pairing and detection of alien DNA transfers in interphase cells of wide crosses involving wheat and rye

The objectives of this study were to determine if biotin-labelled total genomic DNA of rye (Secale cereale L.) could be used to (i) preferentially label rye meiotic chromosomes in triticale and (ii) detect translocation stocks at interphase and/or early prophase by in situ hybridization. Welsh triticale, a wheat-rye segmental amphiploid, and Kavkaz wheat, a wheat-rye translocation were used. The results indicated that labelled chromosomes of rye and unlabelled chromosomes of wheat could be observed throughout all meiotic stages in the triticale. For Kavkaz wheat, the presence of the translocated 1RS chromosome arm of rye was detected at the interphase or very early prophase stage. Rapid assessment of feasibility of gene transfers and detection of alien DNA in somatic cells at the interphase stage by in situ hybridization allows for rapid decision-making and saves time and expense in plant breeding programs.

Hordeum chilense repetitive sequences. Genome characterization using biotinylated probes

A library of random DNA fragment clones of wild barley Hordeum chilense was screened for clones of repeated nucleotide sequences. Five clones were isolated that gave a stronger hybridization signal in colony and dot blot hybridization with total H. chilense DNA in comparison to Triticum aestivum DNA. Clones labelled with biotinylated nucleotides were used as probes to investigate the repeated sequences organization in the H. chilense genome. Tandemly arranged and interspersed sequences have been found, together with homology differences with related sequences present in T. Aestivum, which could allow the differentiation of H. chilense DNA when it is present in wheat. We show that biotin can replace the use of (32)P in preparing repeated sequence probes for Southern and DNA dot blot analyses.

DNA sequence organization in the genomes of three related millet plant species

A major portion of the genomes of three millet species, namely, barn yard millet, fox tail millet and little millet has been shown to consist of interspersed repeat and single copy DNA sequences. The interspersed repetitive DNA sequences are both short (0.15-1.0 kilo base pairs, 62-64% and long (>1.5 kilo base pairs, 36-38%) in barn yard millet and little millet while in fox tail millet, only long interspersed repeats (>1.5 kilo base pairs) are present. The length of the interspersed single copy DNA sequences varies in the range of 1.6-2.6 kilo base pairs in all the three species. The repetitive duplexes isolated after renaturation of 1.5 kilo base pairs and 20 kilo base pairs long DNA fragments exhibit a high thermal stability with Tms either equal to or greater than the corresponding native DNAs. The S1 nuclease resistant repetitive DNA duplexes also are thermally stable and reveal the presence of only 1-2% sequence divergence.The present data on the modes of sequence arrangement in millets substantiates the proposed trend in plants, namely, plants with 1C nuclear DNA content of less than 5 picograms have diverse patterns of sequence organization while those with 1C nuclear DNA content greater than 5 picograms have predominantly a short period interspersion pattern.

Rye cytology, cytogenetics and genetics - current status

Progress in rye karyology is reviewed with respect to chromosome structure, recognition and chromosome nomenclature. Considerable contributions have been brought about by molecular techniques which have even revealed nucleotide sequences of some of the ribosomal DNA. DNA sequence organization correlates with the distribution of major Giemsa C-band regions as well as with N-bands and the binding sites of fluorescent dyes. The several banding patterns permit the classification of rye chromosomes. The increased data and widespread application of banding analysis require a consistent system of chromosome and/or band designation. Therefore, a standard band nomenclature is proposed with reference to the recommendations of the "Paris Conference on Standardization in Human Cytogenetics". In addition, advances in genetics are summarized and discussed. Based on the original accepted standard karyogram and banding patterns of the rye chromosomes, meanwhile, 120 genes determining several characters have been associated with individual chromosomes and/or chromosome arms, including linkage studies for about 19 arrangements. Most results were obtained using wheat-rye addition lines as well as test crosses with defined translocations. Moreover, genetical studies based on appropriate trisomic and telotrisomic material resulted in the localization of 19 genes, including their linkage relationships.

Structure and developmental regulation of a wheat gene encoding the major chlorophyll a/b-binding polypeptide

A genomic clone for a major chlorophyll a/b-binding polypeptide of the light-harvesting complex has been sequenced from wheat. This gene, whAB1.6, encodes a 70-nucleotide 5'-nontranslated spacer, a 34-amino-acid NH2-terminal extension, i.e., the transit peptide, and a mature coding protein of 232 amino acid residues. The exact molecular weight of the precursor polypeptide is 28,560. The transit peptide is basic and is rich in serines. No intervening sequences are found in this gene. The transcription start site of the whAB1.6 gene occurs at AAAC as determined by S1 nuclease analysis. Putative regulatory sequences occur upstream of the gene at -25 (TTTAAATA) and at -72 (CCAACCA). Northern blots show a single RNA species estimated to be 1,100 nucleotides. Heterogeneity of the RNA population is demonstrated in S1 nuclease analyses with a 5'-end-labeled fragment that extends 191 nucleotides into the mature protein coding sequence. At least seven different transcripts can be recognized. The highest levels of RNA transcribed from the whAB1.6 gene are found in the basal segments of the wheat leaf, whereas other chlorophyll a/b-binding transcripts in the cell show a different pattern of abundance. As a control, we show that roots do not contain chlorophyll a/b-binding RNA. The most abundant RNA species shows an interrupted homology with the whAB1.6 gene at the start of the mature protein coding sequence; another species shows homology beginning at the start of the transit peptide and does not include the nontranslated region. Chlorophyll a/b-binding polypeptides accumulate toward the tip of the leaf as shown by Western blot analysis of total thylakoid proteins.

Structure and developmental regulation of a wheat gene encoding the major chlorophyll a/b-binding polypeptide

A genomic clone for a major chlorophyll a/b-binding polypeptide of the light-harvesting complex has been sequenced from wheat. This gene, whAB1.6, encodes a 70-nucleotide 5'-nontranslated spacer, a 34-amino-acid NH2-terminal extension, i.e., the transit peptide, and a mature coding protein of 232 amino acid residues. The exact molecular weight of the precursor polypeptide is 28,560. The transit peptide is basic and is rich in serines. No intervening sequences are found in this gene. The transcription start site of the whAB1.6 gene occurs at AAAC as determined by S1 nuclease analysis. Putative regulatory sequences occur upstream of the gene at -25 (TTTAAATA) and at -72 (CCAACCA). Northern blots show a single RNA species estimated to be 1,100 nucleotides. Heterogeneity of the RNA population is demonstrated in S1 nuclease analyses with a 5'-end-labeled fragment that extends 191 nucleotides into the mature protein coding sequence. At least seven different transcripts can be recognized. The highest levels of RNA transcribed from the whAB1.6 gene are found in the basal segments of the wheat leaf, whereas other chlorophyll a/b-binding transcripts in the cell show a different pattern of abundance. As a control, we show that roots do not contain chlorophyll a/b-binding RNA. The most abundant RNA species shows an interrupted homology with the whAB1.6 gene at the start of the mature protein coding sequence; another species shows homology beginning at the start of the transit peptide and does not include the nontranslated region. Chlorophyll a/b-binding polypeptides accumulate toward the tip of the leaf as shown by Western blot analysis of total thylakoid proteins.

Organization and evolution of repeated DNA sequences in closely related plant genomes

In common with many other eukaryotic species, the genomes of species in the genus Allium contain a high proportion of repeated DNA sequences, which may be implicated in the considerable differences in genome size that are seen between even very closely related species. The gross organization of repetitive sequences within the genome of Allium sativum and of some other related species has been investigated using DNA/DNA hybridization studies. Such studies show that there has been much modulation in the amounts of different repeated DNA families during the evolution of the genus Allium; these repetitive elements are interspersed in all species with sequences of low repetition. The organization and distribution of one particular repetitive family within the genus has been examined using a cloned hybridization probe. Hybridization of this probe to DNA from related genomes reveals that this element is present in all Allium species examined, but with large-scale modulation of its abundance, and some considerable changes in its sequence environment. The evolution of such genome-specific arrangements of common repetitive elements and the possible mechanisms by which they might be maintained are discussed.

Repeated deoxyribonucleic acid clusters in the chicken genome contain homologous sequence elements in scrambled order

Part of the repeated deoxyribonucleic acid (DNA) in the chicken genome had a clustered organization. The following description of clustered repeated sequences is derived both from analysis of DNA segments cloned in lambda and from hybridization of individual cloned sequences to Southern blots of restricted total DNA. A cluster usually exceeds 20 kbp in length and consists principally, if not entirely, or repetitive DNA. Each cluster contains one cope of several different repeated sequences. The individual sequences occur several hundred times in the genome, but only once per cluster. Many of the clusters contain the same assortment of sequences but in scrambled order. In the genome, those repeated sequences that are elements of clusters occur mainly within the clustered context and are seldom, if ever, found as isolated elements flanked by nonrepeated DNA. These aspects of cluster organization suggest that the clustered sequences undergo limited rearrangement, maintaining the associations within clusters but allowing variability of sequence arrangement from cluster to cluster. The clusters that occupy the cloned DNA segments together represent at least 10% of the repetitive DNA of the chicken.

Evolutionary sequence divergence within repeated DNA families of higher plant genomes. II. Analysis of thermal denaturation

An assay based on derivative analysis of thermal denaturation (melting) behavior of reassociated DNA was developed in an attempt to characterize the sequence relationships in repeated DNA families according to the homogeneous or heterogeneous models of Bendich and Anderson (1977). The validity of the technique was confirmed by the use of deaminated Escherichia coli DNA models for repetitive families. The melting data for DNA reassociated at two different temperatures provided strong evidence that Pisum sativum repeated families are mostly heterogeneous, while homogeneous families predominate in Vigna radiata. These findings, together with other differences between the two genomes, suggest that the rate of sequence amplification has been higher in the evolutionary history of Pisum DNA. A general trend seems to exist for high amplification rates in large, highly repetitive plant genomes such as Pisum and lower rates in smaller plant genomes such as Vigna, as well as in the generally smaller, less repetitive genomes of most animal species.

DNA sequence organization in the flax genome

The complexity of the flax genome has been determined by reassociation kinetics. The total complexity of one constituent genome was 3.5 . 10(8) nucleotide pairs. The single copy sequences comprised 44% of the genome and showed a long period interspersion pattern with the repetitive sequences. The repetitive sequences occurred in clusters which stretched for at least 10 000 base pairs. Within these clusters the individual repetitive elements were about 650 base pairs. These elements themselves showed little interspersion of different frequency classes in lengths less than 3000 base pairs. The repetitive sequence duplexes formed on reassociation, except for the satellite DNA, showed a high thermal stability. The fold-back DNA comprised 1% of the total genome, and was itself clustered in a small fraction of the genome.

Sequence organization of animal nuclear DNA

Animal nuclear genomes contain DNA sequences of various degrees of repetition. These sequences are organized in highly ordered fashions; repetitive and nonrepetitive sequences either alternate in short periods, i.e., short [0.2-0.4 kilobases (kb) long] repeats are flanked by nonrepetitive sequences less than 2 kb long, or in longer periods, with repetitive and/or nonrepetitive sequences extending for several kilobases. There are two main categories of genome organization, namely those exhibiting short-period interspersion and those that do not. There are arguments for and against a regulatory role of short interspersed repetitive sequences. Besides the merely 'statistical' kinetic approach by conventional reassociation kinetics, sequence organization has been studied by restriction endonuclease mapping and nucleotide sequencing. Such studies have revealed some general features of the organization of the eukaryotic gene and its transcripts, namely possible 'promoters', 'leaders', 'introns', 'exons', 'flanking sequences', 'caps', ribosome-binding sites, and poly(A) sequences. This paper discusses how these elements of a gene might serve regulatory roles in its expression.

Deoxyribonucleic acid sequence organization in the genome of the dinoflagellate Crypthecodinium cohnii

Details of the general DNA sequence organization in the dinoflagellate Crypthecodinium cohnii have been obtained by using hydroxylapatite binding experiments, S1 nuclease digestion .and electron microscopy of reassociated DNA. It has been found that roughly half of the genome is made up of unique sequences interspersed with repeated sequence elements with a period of approximately 600 nucleotides. This class represents roughly 95% of the total number of interspersed unique elements in the genome. The remaining 5% are uninterrupted by repeated sequences for at least 4000 nucleotide pairs. The interspersed repeated elements are narrowly distributed in length with 80% under 300 nucleotide pairs in length. About half of the repeated DNA (20-30% of the genome) is not interspersed among unique sequences. The close spacing of the short repeats interspersed throughout much of the genome is consistent with the occurrence of the huge network structures observed in the electron microscope for low Cot reassociation of moderately long fragments. An unusual class of heteroduplexes was detected in the electron microscope which is believed to derive from the reassociation of repeated sequences from different families which are frequently found adjacent to one another in different locations in the genome. The occurrence of this novel arrangement of repeated sequences may reflect the unusual organization of the dinoflagellate nucleus. However, in most respects the sequence arrangement in this unicellular alga is very typical of higher plants and animals.

Characterization of long and short repetitive sequences in the sea urchin genome

Long and short repetitive sequences were purified from the DNA of Paracentrotus lividus under conditions designed to optimize the yield of complete, end to end sequences. Double-stranded long repeat DNA prepared in this manner ranged in length from approximately 3000 to 15 000 nucleotide pairs with average sizes of approximately 6000 base pairs. In the electron microscope, long repeat DNA was observed to possess continuous sequences that often appeared to be terminated by one or more loops and/or fold backs. Long repeat DNA sequences, resheared to 300 base pairs, were found to have an average melting point identical to that for sheared native DNA. Thus, the reassociated duplexes of long repetitive DNA seem to possess very few mismatched base pairs. Reassociation kinetic analyses indicate that the majority of the long repeat sequences are reiterated only 4--7 times per haploid amount of DNA. Melt-reassociation analyses of short repetitive DNA, at several criteria, support the previously held concept that these sequences belong the sets or families of sequences which are inexact copies of one another. Our studies also support hypotheses suggesting that short repetitive sequences belong to families which may have arisen via distinct salttatory events. The relationships between long and short repetitive DNA sequences are considered with respect to widely held concepts of their sequence organization, evolution, and possible functions within eucaryotic genomes. A model for the possible organization of short repeats within long repetitive DNA sequences is also presented.

The organization of a nuclear DNA sequence from a higher plant: molecular cloning and characterization of soybean ribosomal DNA

The recombinant DNA vector, lambda Charon 4A, was used to construct a library of DNA sequences from the genomic DNA of soybean (Glycine max). To define the organization of ribosomal DNA (rDNA) in the soybean genome, clones containing sequences complementary to both 17S and 25S rRNA have been isolated from this library and used in conjunction with Southern blot hybridization. The rRNA genes are tandemly reiterated with a relatively small unit repeat length of 7.8 kb. There is no heterogeneity in the length of the rDNA repeat units although they display limited differences in either base sequence or pattern of methylation. The cloned rDNA sequences are shown to comprise the entire repeat unit and have been used to obtain a detailed restriction map as well as an approximate transcription map of soybean rRNA genes. The cloning of rDNA from soybean suggests that recombinant DNA techniques can be successfully applied to the genomic DNA of higher plants despite the high degree of methylation exhibited by plant DNA.

The evolution of the long and short repetitive DNA sequences in sea urchins

The rates of evolution of purified long and short repetitive DNA sequences were examined by hybridisation analysis between the DNAs from several species of sea urchins. We find that the rates of nucleotide substitution are very comparable within mutually retained sequences for the two classes of repetitive DNA. The loss of hybridisable sequences between species also occurs at similar rates among both the short and long repetitive DNA sequences. Between species that separated less than 50 million years ago, hybridisable short repetitive sequences are lost all through the spectrum of reiteration frequencies. The long repeats contain a few sequences which are highly conserved within all of the species examined, and which amount to approximately 1% of the total genome. The short repetitive class, on the other hand, does not seem to contain any such highly conserved elements. The long repetitive sequences internally appear to contain short 'units' of reiteration, which may comprise families within the long repetitive class. We find no evidence to indicate that the majority of long and short repetitive sequences evolve by different mechanisms or at different rates.