Pl. Syst. Evol. 213:233-250 (1998) Plant Systematics and Evolution © Springer-Verlag I998 Printed in Austria Molecular phylogenetic relationships in Aveneae (Poaceae) species and other grasses as inferred from ITS1 and ITS2 rDNA sequences B. GREBENSTEIN,M. RöSER, W. SAUER,and V. HEMLEBEN Received May 2, 1997; in revised version August 1, 1997 Key words: Poaceae, Aveneae, Trisetum, Koeleria, Helictotrichon, Avena. - Ribosomal DNA, ITS region, GC content, phylogenetics, systematics. Abstraet: A phylogenetic analysis was conducted on sequences of the internal transcribed spacer (ITS) region of nuclear ribosomal DNA in 23 species ofAveneae (Poaceae subfam. Pooideaae). These sequences of Helictotrichon spp., Arrhenatherum elatius, Avena spp., Trisetum spp., Koeleria spp., Holcus lanatus, Alopecurus vaginatus together with published ITS sequences of further Aveneae, Poeae, Triticeae, and Bromeae were analysed by the neighbor-joining distance method to assess the molecular phylogenetic relationship in perennial and annual Aveneae. The results suggest unexpectedly close affinities of the agronomically important genus Avena to comparatively small-flowered taxa of Aveneae. Genus Arrhenatherum and small-flowered subgenera of Helictotrichon are close extant relatives. The large genus Helictotrichon is para- if not polyphyletic, only its subgenera are monophyletic. Trisetum is clearly separated from Helictotrichon and forms together with Koeleria and perhaps others a monophyletic lineage which is characterised by a conspicuous 9-bp deletion in ITS1. The impact of the ITS data on the delineation of some genera and subtribes of Aveneae and on the recognition of their biogeographical and ecological patterns is outlined. The region of the internal transcribed spacers ITS1 and ITS2 of the 18S-25S nuclear ribosomal DNA has been established as a useful marker to decipher relationships among plants (HEMLEBEN 1993, BALDWIN & al. 1995). Within the family Poaceae, only a few taxa such as species of the genus Zea (BUCKLER & HOLTSVORD 1996a, b), the tribe Triticeae (HsIAO & al. 1995a) and further members of subfam. Pooideae (HSIAO & al. 1994, 1995b) were investigated using this molecular marker. Both ITS1 and ITS2 appeared to be variable enough to differentiate between closely related species. Variation between ITS sequences is mostly due to point mutations, less to insertions/deletions (indels) of nucleotide stretches. In this study we examined the ITS region of different taxa of the tribe Aveneae to analyse molecular phylogenetic relations among some genera that are assumed 234 B. GREBENSTEIN& al.: to be closely related to the agronomically important genus Avena. To date, the closest relatives of Arena species are still not known, and our goal was to clarify the unclear relationships within the subtribe Aveninae, where the genera Arena, Arrhenatherum and Helictotrichon have been grouped together according to morphological and anatomical features. In general, these features are very suitable to resolve phylogenetic relationships, but in these taxa they are partly incompletely known or ambiguous. Some characters that had traditionally been employed in the taxonomy of the Aveneae actually rest on parallelism within disparate groups (RöSER 1989, 1996). Particular attention was paid to the genus Helictotrichon, the largest genus of Aveneae with c. 100 species. It is worldwide distributed and shows a considerable morphological and karyological diversity. The systematic position of the genus Helictotrichon within the tribe Aveneae and its close relationship to the genera Arena, Arrhenatherum and Trisetum had never been questioned. However, the delimitation of genera or infrageneric taxa has always been a matter of debate (BENTHAM 1883, V~ERHAPPF~R 1914, HOLt;B 1958, CLAYTON & RENVOIZE 1986). Our molecular sequencing data presented here thus permit an independent evaluation of this subdivision and provide new insight into the evolutionary history of this group. To assess molecular phylogenetic relationships of the tribe Aveneae, ITS sequences of grasses from the tribes Poeae, Bromeae and the agronomically important Triticeae, all belonging to the subfam. Pooideae, were included for comparison. Materials and methods Plant material. Living plant material of wild perennial oats and other grasses of subfam. Pooideae (Table 1; GREBENSXEIN& al. 1996) were collected in their natural habitats and were grown in pots in the greenhouse (23 °C in summer, 17 °C in winter). DNA isolation, amplification and sequencing. Isolation of plant total DNA was done according to GREBENSTEIN & al. (1995). Two primers designed for Cucurbita pepo (Cucurbitaceae) were used for PCR amplification and cloning into the Eco RI site of pUC 18 (JOBST & al. 1998). Primer 18S-5.8S (Y-GCGAGAATTCCACTGAACCT-3') is complementary to 18S rDNA sequence near the ITS1 border, and pfimer 25S-5.8S (5 IACGAATTCCCTCCGCTTATTGATATGCTTA-31) anneals to 25S rDNA near the ITS2 border. These two primers flank the entire ITS region, which includes the 5.8S rDNA that separates the ITS1 and ITS2 regions. PCR was carried out in a Mini CyclerT M (MJ Research). The 50 gl reaction mixture assay contained 100 ng of template DNA, 20 gM of each primer, 5 ~tl of 10 mM dNTP mixture, 5 pl of 10x Tfl-Polymerase buffer and 1 unit Tfl-Polymerase (Biozym). The reaction mixture was overlaid with a drop of paraffin and preheated to 95 °C for 5 min. The thermal cycle program was run for 30 cycles (95 °C 45 s; 55 °C 45 s; 72 °C 60 s), followed by a 10 min final extension at 72 °C. The reaction was checked in a 1% agarose gel. PCR products were purified with phenol and precipitated with isopropanol. Restriction endonuclease digest with EcoRI was carried out for 1 h. After ethanol precipitation the fragments were ligated into the phosphatase-treated EcoRI cloning site of pUC18 plasmid, and the ligation products were transformed into E. coli XLl-blue MRF' competent cells. Several recombinant clones from each transformation which contained the ITS inserts were selected by colony hybridisation using a digoxigenin- Molecular systematics o f Aveneae 235 o © °° ~ ô~ ~'~ .~ o 0.) 7- ~~ c~ o~ .~ bh 0 ~.~~ ~D ùC "~ "~ o ~D O o cq q ~D O O ~q g -d '~ ° ° I o° cq o~ o O tr3 ~~ < ~~~ o m o © ~D 0..i ù. õ o o o eh ù~ ~õ u et) .=~ ~ o --I v © Z .~ ~D ~z :õ © { 10 :8 " ~z = z ~ .~ < ~ ~~ Z g~ © ~A ~m # ..-q ~q ~,4 © ~ m . ù~ o cq < (D © Z zg ~ "2'. ~-~ "~ = , b~ 236 B. GREBENSTEIN & al.: ..= ~ e ù~ > g.~ o c~ gg ,g © .ä ù~ © ù-:, c~ ù o < ~g .e b~ ~ :8 c~ ~q ù~ ~ =~ ~ g1¢3 © ù~ ~a "~ ~ ~ .~ z~~ B "q g" u ~ e'~ ù~ .gg< ~g d ~ ~.~~ -q ~g ;o © N < r,~ ~5 © ',= I~ ~ .~~ ~ ~ ~ ù~ ~ C/~ © m .ô ~~~ ~~~ ~~~< >-, C~ ùa 2 ~ 2 ä r-i ca ¢q © o .~ .~~ .~-~~ © © ~a ù~ © .< o & "~~~~ © © ~ ~ .~ 'r. ~ ~ .~. ,-a ,-a Molecular systematics of Aveneae 237 labelled ITS PCR product from Helictotrichon versicolor DNA. Plasmid DNA isolation was done using the QIAwell-Kit (Diagen); sequencing reaction of 2-7 recombinant clones of each species under investigation was performed with the AutoREAD sequencing kit and an ALF sequencing unit (Pharmacia). The entire sequence of ITS1 and ITS2 regions (without 5.8S rDNA sequence) were aligned using the "Align Plus/sequence alignment program" (MYERS& MILLER 1988) and refined by eye in Word for Windows. Phylogenetic reconstruction was performed using the Jukes-Cantor correction method (JUKES & CANTOR 1969) and the neighbor-joining distance matrix method of the 'Treecon 1.2' package (VAN DE PEER & DE WACHTER 1993). Computation of evolutionary distances with other correction models (TAJIAMA & NEI, KIMURA -- two parameter) and cluster analysis methods for tree construction resulted in similar topologies of the evolutionary trees (data not shown; see VAN DE PEER & DE WAC•TER 1993 for details of inferring evolutionary trees and the computation of distances between sequences). Bootstrap analysis to test the reliability of branches in evolutionary trees were calculated according to FELSENSTEIN (1985) in Treecon 1.2. Nueleotide sequenee data. The nucleotide sequence data obtained in this study appear in the EMBL, GenBank and DDBJ Nucleotide Sequence Databases under the accession numbers Z96812-Z96922 (ITS1 with even numbers) and Z96813-Z96923 (ITS2 with uneven numbers). Results and discussion Sequence analysis. ITS1 and ITS2 sequences of 58 independent clones from 23 species of Aveneae (Tables 1, 2) were analysed to determine the variation in length and G+C-content, the infraspecific homology, and to study the molecular phylogenetic relationships among this group. The organisation of the nuclear rDNA repeats of the Aveneae and further taxa of grasses are similar to that of other angiosperms (HAMBY& ZIMMER 1988, BALDWlN 1992). The entire ITS1 and ITS2 sequence investigated here (Table 2) without 5.8S rDNA, ranged from 420 bp (Trisetum turcicum) to 439 bp (Helictotrichon pubescens). ITS1 region varies between 203 bp (Trisetum turcicum) and 225 bp (Helictotrichon pubescens). The length of ITS2 ranges from 211 bp (Avena sativa) to 220 bp (Helictotrichon convolutum). The G+C content of the entire regions averages 63%, ranging from 58.8% (Avena sativa) to 66.6% (Helictotrichon sarracenorum). The G+C content of ITS1 averages 63.7%, ranging from 56.9% (Avena sativa) to 66.6% (several species of Helictotrichon subg. Helictotrichon). In ITS2, the G+C content shows values between 60.2% (Holcus lanatus) and 66.5% (species of Helictotrichon subg. Helictotrichon). In subg. Helictotrichon the G+C contents of ITS1 and ITS2 sequences are almost corresponding with differences up to 1.6%, whereas in the other subgenera of Helictotrichon the G+C content of ITS1 is slightly, but constantly lower (2.84.4%) than the G+C content of ITS2 (Table 2). These results are congruent with the observation that the G+C-content of ITS 1 and ITS2 regions in plants and animals are almost balanced (TORRES& al. 1990). SA5INAS & al. (1988) showed by reassociation kinetics of single-stranded DNA that grasses growing in arid regions have on average a higher G+C content than plants from temperate areas, and an adaptive significance of the G+C content was suggested (cf. BERNARDI ~ al. 1985, 1988). 238 B. GREBENSTEIN & al.: Table 2. ITS sequence lengths, base pair composition, infraspecific variation, and chromosome numbers in Aveneae and species of the tribes Poeae, Triticeae, and Bromeae. nd not determined in the present study; ni not indicated by the authors Taxon Chrorno- ITS 1 region some number Length % Ge 2n [bp] ITS2 region ITS I+ITS2 region Length [bp] % GC Length [bp] %Ge Infrasp. rar. [%] Helictotrichon subg. Helictotrichon H. sarracenorum 14+1B H. convolutum 14 14. filifolium subsp, filifolium 84 H. cantabricum 84 H. sempervirens 42+1B 216-217 216-217 216-217 218-219 217-218 66.6 66.6 66.6 66.6 65.3 213-214 214-220 214-215 214-217 214-215 66.5 65.9 66.5 65.0 65.3 429-431 430-437 430-432 432-436 431-433 66.6 66.2 66.5 65.8 65.3 1 2 3 3 0 H. subg. Tricholemma H. jahandiezii 28 221 63.0 215-216 65.8 436-437 64.4 1 H. H. H. H. H. H. H. H. H. 14 14 42 14 72 126 28 14 217-219 218 218 219 218 218 217-219 216-217 60.8 59.4 60.7 60.3 61.0 61.0 61.2 61.0 217 215-216 215 216 216 216-217 216 214-216 64.3 63.8 65.1 64.4 64.1 64.0 64.1 65.3 434-436 62.5 433-434 61.6 433 63.0 435 62.3 434 62.6 434-435 62.4 433-435 62.4 430-433 63.2 1 2 nd nd 1 1 2 1 14 218-225 59.9 213-214 63.6 431-439 61.7 2 Arrhenatherum elatius Avena longiglumis I A. sativa Trisetumflavescens T. turcicum Koeleria digorica K. pyramidata Briza minor 1'2 Deschampsia cespitosa a Holcus lanatus Phalaris truncata 1 Alopecurus vaginatus 28 14 42 nd 28 28 nd ni ni nd ni 28 215-219 218 215-218 207-209 203-208 208-209 208-209 216 217 216-219 232 217-220 60.4 56.9 56.9 66.1 65.6 65.5 65.2 56.0 61.3 60.6 57.3 61.1 214-215 213 211-213 217 217 217-219 216-217 214 216 216 203 212-213 61.2 62.4 61.4 63.6 63.3 63.2 63.9 61.2 64.4 60.2 60.1 61.4 429-434 431 426-431 424-426 420-425 425-428 424-426 430 433 432-435 435 429-433 60.6 59.6 58.8 64.1 64.6 64.4 64.3 58.6 62.8 60.4 58.6 61.3 0 ni 2 1 2 2 1 ni ni 3 ni 1 Poeae Festuca mairei I Loliumperenne 1 Dactylis glomerata a ni ni ni 219 219 211 60.3 58.0 60.7 214 214 215 61.2 64.0 60.0 433 433 427 60.7 61.0 60.2 ni ni ni Triticeae Secale cereale I 14 221 61.1 216 60.6 437 60.9 ni Bromeae Bromus inermis x ni 217 54.8 215 59.1 432 56.9 ni subg. Pratavenastrum bromoides subsp, bromoides compressum hackelii aetolicum armeniacum pratense s. 1. blaui subsp, blaui versicolor subsp, versicolor H. subg. Pubavenastrum H. pubescens Further Aveneae IITS data according to Hsino & al. (1994, 1995a, b) 2Tribal position as suggested by SOREN6 & al. (!990), HsIAO & al. (1995b), and this paper Moleculär systematics of Aveneae 239 Our analysis of the G+C content of the ITS 1/ITS2 sequences in Helictotrichon and other Aveneae fits these observations since the highest G+C contents occur in throughout strongly xeromorphic species of the subg. Helictotrichon. Additionally, the most thermophilic, drought-resistent and strictly Mediterranean distributed species of this subgenus (H. sarracenorum, H. convolutum, H. filifolium) have a slightly higher G+C content than the species of the thermophilic vegetation at the edges of the Mediterranean (Pyreneo-Cantabrian H. cantabricum, H. sempervirens in the SW Alps; cf. Tables 1, 2). The high values of the subg. Helictotrichon are followed, in this sequence, by lower ones in the relic subg. Tricholemma (H. jahandiezii with slightly succulent habit occurring in dry mountain garigues of the Central Moyen Atlas, Morocco), the subg. Pratavenastrum (with variously adapted species), and finally the monotypic subg. Pubavenastrum where the constantly mesomorphic species H. pubescens is exceptionally widely distributed in moderately dry to varying moist grassland vegetation from western and northern Europe through East Siberia (Table 1). The same pattern seems to hold true in the other Aveneae studied, although they have been less extensively sampled. Starting with species of Trisetum (64.1-64.6%), of Koeleria (64.3-64.4%), and Alopecurus vaginatus (61.3%), which are colonisers of dry rocky heath vegetation of Europe and southwestern Asia, they show subsequently decreasing G+C contents in species of moderately dry grasslands (60.6% in Arrhenatherum elatius which ecologically and chorologically resembles Helictotrichon pubescens), to species of habitats with almost continuous watet supply: 60.4% is found in Holcus lanatus (seasonally inundated wetlands of Europe) and 58.6% in Phalaris truncata which is characteristic of damp places of the western and southern Mediterranean. All of these species of Aveneae are perennials. Consistently low values of 58.6-59.6% are characteristic of the annual species of Aveneae, namely Avena sativa, A. longiglumis, and Briza minor which according to strong molecular evidence is part of this tribe (Fig. 2; SO~EN~ & al. 1990, HSIAO & al. 1995b). These annual Aveneae are colonisers of habitats with strong seasonality, usually caused by a severe drought period. In contrast to the perennials, they do not show any elaborate morphological adaptations to this periodic drought (RösER 1997), but rely completely on their capability to accomplish germination, vegetative growth, flowering, fruiting and seed disperal within a short time. With respect to this absence of conspicuous morphological or anatomical adaptations to drought, the annuals resemble the mesomorphic perennials with similarly low G+C contents. A correlation of annual life form with special demands on DNA composition such as suggested by our data on the ITS regions therefore needs to be tested in a broader range of organisms. Different constraints on biological features of perennials and annuals are apparent and well known, for example in the respective ranges of total nuclear DNA amounts (BENNETT 1972, 1987) or the particular breeding systems (CoN~oR 1979, 1987). Moleeular phylogenetie analysis. The alignment of ITS1/ITS2 sequences from 32 species resulted in 492 characters. In the phylogenetic reconstruction of the neighbor-joining tree 113 sites were informative. S u b t r i b e A v e n i n a e . Detailed information obtained in many species of the genus Helictotrichon and its close atlies Avena and Arrhenatherum is shown in the neighbor-joining dendrogram with Lolium perenne of the tribe Poeae as outgroup 240 B. GREBENSTEIN & al.: Distance 0.1 4 71~-Helictotrichon filifolium 3 L__ Helictotrichon filifolium 1 511 L_ Helictotrichon filifolium 2 ! l ~ Helictotrichon convolutum 1 45~ .... -750'2~Helictotrichon convolutum 3 ~]~ Helictotrichon Helicl convolutum 2 67~98~---- Helictotrichon sempervirens 2 II L_Helictotrichon sempervirens 1 B ~ Helictotrichon sarracenorum 2 B ~ - Helictotrichon sarracenorum 1 5 ~ H H : c t o t r i c h o n cantabricum 2 f I ~ Helictotrichon cantabricum 1 _ ~00F Helictotrichon jahandiezii 3 66~Helictotrichon jahandiezii 2 L_ Helictotrichon jahandiezii 1 100[ Arrhenatherum elatius 1 ~ Arrhenatherum elafius 3 69 ~ Arrhenatherum elatius 2 [ - - Arena longiglumis* 10018 1 ~ Arena sativa 5 -54j u Avena sativa 4 6~5~-~Avenasativa 3 5'8_.[---Avena sativa 2 76t Avena sativa 6 Avena sativa 1 ___L~HHelictotrichon versicoior 2 -elictotrichon versicolor 1 Helictotrichon hromoides 2 57t ~ - Helictotrichon bromoides 1 100 52~L_Helictotrichon blaui 1 L~_ Helictotrichon hackelii 1 100 B L Helictotrichon compressum 2 Helictotrichon compressum I 59~ Helictotrichon armeniacum 2 19~ ~ - Helictotrichon aetolicum 1 F- Helictotrichon pratense 1 43~[ Helictotrichon pratense 2 6~ Helictotrichon blaui 2 Helictotrichon armeniacum I _ F- Helictotrichon pubescens 1 i°BR 7 - ~ ' Helictotrichon pubescens 4 46~ Helictotrichon pubescens 3 a Helictotrichon pubescens 2 Lo]ium perenne* B I subg. Helictotrichon ] subg. Tricholemma subg. Pratavenastrum ] subg. Pubavenastrum Fig. 1. Neighbor-joining tree inferred from rDNA ITS sequences in 'core genera' of Aveneae subtribe Aveninae analysed with the distance matrix method. Numbers above branches are bootstrap values. Branch lengths are proportional to distance. Outgroup: Lolium perenne (Poeae). Asterisk designates sequence data taken from HSIAO & al. (1994, 1995b) Molecular systematics of Aveneae 241 (Fig. 1). The diagram resolves the large genus Helictotrichon as paraphyletic, but its subgenera Helictotrichon, Tricholemma, Pratavenastrum and Pubavenastrum (cf. RöSER 1989) as monophyletic. Two of rhein, subg. Helictotrichon and subg. Tricholemma, belong together with the genera Arrhenatherum and Arena to a lineage separate from the subgenera Pratavenastrum and Pubavenastrum. Within the species-rich subgg. Helictotrichon and Pratavenastrum it is not possible to resolve their subordinate subgroups (H. sarracenorum-, H. parlatorei group, and H. bromoides-, H. marginatum-, H. versicolor-, H. aetolicum-, H. blaui group) with ITS sequences as molecular markers. Infraspecific variability and evolutionary distances detected between several independent ITS repeats within one species (Table 2) is frequently larger than their evolutionary distances to sequences of species from other subgroups of the same subgenus (Fig. 1). These subgroups, however, are characterised by particular anatomical and morphological features and usually show consistent patterns of distribution (RöSER 1989, 1996). In the xeromorphic subg. Helictotrichon, the strictly Mediten'anean H. sarracenorum group represented by the diploids H. sarracenorum (2n=14+lB) and H. convolutum (2n=14) and the dodekaploids H. cantabricum and H. filifolium Distance0.1 4~ q H. subg. Helictotrichon 98 ~ H. subg. Tricholemma ~ ~ 0~90~Koeleria digorica L_ Trisetum flavescens oeleria pyramidata Trisetum turcicum Arrhenatherum elatius AÄvnê2ä~älg~:mis* Aveneae Phalaris truncata* Briza minor* Holcus lanatus H. subg. Pubavenastrum Alopecurus vaginatus Deschampsia cespitosa* 100 100 H. subg. Pratavenastrum Dactylis giomerata* 9~1 Loliumperenne* - - - Poeae Festuca mairei Bromus inermis* Secale eereale* Bromeae Triticeae Fig. 2. Phylogenetic tree inferred from ITS sequences of 15 genera of tribes Aveneae and Poeae generated by the neighbor-joining distance matrix method. Numbers above branches are bootstrap values. Branch lengths are proportional to distance. Outgroups: Bromeae (Bromus inermis) and Triticeae (Secale cereale). Asterisk designates sequence data taken from HSlAO& al. (1994, 1995b) 242 B. GREBENSTEIN& al.: subsp, filifolium (2n--84) does not differ from hexaploid H. sempervirens (2n=42+lB) which is a member of the H. parlatorei group, a florally specialised group of Helictotfichons endemic to the Alps (Fig. 1, cf. Table 1). Infraspecific variability of 3% in highly polyploid H. cantabricum and H. filifolium suggests the presence of at least two slightly different types of ITS repeats. However, in diploid H. convolutum the infraspecific variability reaches 2% (Table 2). A presence of different ITS types within the same species was recently reported, for example, in Pinus where multiple length variants of ITS1 occur (MAGGINI • BALDASSINI 1995), in hexaploid hybrid Krigia montana (Asteraceae) which contains polymorphic ITS sequences (KtM & JANSEY 1994), and North American taxa of Amelanchier (Rosaceae) which show extensive ITS sequence polymorphism within individuals (CAMPBELL & al. 1997). For these taxa polyploidisation and/of agamospermy were considered as possible reasons of polymorphism of ITS sequences. In the species of Helictotrichon, different ITS types do not differ significantly by length, but solely by nucleotide substitutions, so definite conclusions about the genornic constitution of the polyploids presently cannot be drawn from these ITS data. An analysis of more independent clones from the species in question would be required. The small endemic North African subg. Tricholemma comprises the Moyen Atlasic H. jahandiezii and H. breviaristatum. The latter is known ffom a single mountain range in the Algerian Hauts Plateanx, but could not be recollected since 1882 and therefore was not at hand for this study. Based on conspicuous characters of leaf architecture subg. Tricholemma was frequently combined with subgg. Pratavenastrum and Pubavenastrum in one genus (viz. Avenochloa or Avenula) different from Helictotrichon (cf. HOLUB 1962, 1976) which was subsequently adopted in various regional systematic treatments or floras (KERöU~LEY 1975, ROMERO ZARCO 1984, FREY 1991). This genus concept, however, was revised in view of more subtle floral characters which are exclusively found in subgg. Tricholemma and Helictotrichon, suggesting that the previously emphasized characters of vegetative morphology would imply some degree of homoplasy (RöSER 1989, 1996). The neighbor-joining tree as inferred by ITS sequencing data supports that subg. Tricholemma with the analysed species H. jahandiezii (2n=28) is a real sistergroup of subg. Helictotrichon. The subgg. Pratavenastrum and Pubavenastrum are more distantly related to these subgenera (Fig. 1). Five major groups of taxa of the widespread and richly evolved holarctic subg. Pratavenastrum (Eurasia, North Africa, 1 species in North America) were investigated (Tables 1, 2). The circum-Mediterranean Helictotrichon bromoides group (2x-18x; RöSER unpubl.) is represented by H. bromoides subsp, bromoides, a widely distributed and constantly diploid taxon (2n--14) of the western Mediterranean. In the similarly distributed though ecologically strongly deviant H. marginatum group (2x-6x), diploid H. compressum (2n=14), distributed in 'forest-steppes' of SE Europe and SW Asia, and the geographically widely separated hexaploid H. hackelii (2n=42) which is endemic to the driest coastal regions of SW Portugal were studied. ITS data on the W Eurasian orophytic H. versicolor group (2x-6x) refer to diploid (2n=14) H. versicolor subsp, versicolor. 18x H. pratense (2n=126) and tetraploid H. blaui subsp, blaui (2n----28) are species of the H. blaui group. This usually highly polyploid group of species is widespread Molecular systematics of Aveneae 243 in the more humid areas of Central and Western Europe (except for Ireland), southern Scandinavia, and the European part of the former Soviet Union. It has centres of species diversity in the Pyrenees, the Alps and the mountains of eastern Central and southeastern Europe. It reaches the Mediterranean only peripherically, namely in the northern Iberian Peninsula and the northern Balkan Peninsula. In these regions the lowest ploidy levels of this species group are found: 12x in Pyreneo-Cantabrian species and 4x in the Illyrian endemic H. blaui subsp, blaui which is covered in this study. A further species group, the Il. aetolicum group (LANGE 1995), consists of the endemic Balkanic orophyte H. aetolicum (2n=14) and the rare Irano-Anatolian H. armeniacum (2n=72). None of these five morphologically, ecologically and geographically well-defined species groups of subg. Pratavenastrum is convincingly resolved in the dendrogram of the ITS sequence data (Fig. 1), apparently for the same reasons as found in the subg. Helictotrichon (cf. above). In H. blaui two different ITS repeat types are grouping into different minor branches. One type (no, 1) shows affinity to the 1t. bromoides/ H. compressum lineage, the other (no. 2) to the H. armeniacum/H, pratense lineage. This might indicate a possible alloploid origin of this 4x species from parents with different genomes (cf. also H. armeniacum repeat types no. 1 and no. 2), but further examinations, for example, by genomic in situ hybridisation (GISH) is required. The monotypic subg. Pubavenastrum with the diploid species H. pubescens (2n=14) is clearly separated from the other subgenera of Helictotrichon (Fig. 1) which seems to be in accordance with its morphologically isolated position (RöSER 1989), since the only character considered previously to indicate close relations to subgg. Pratavenastrum and Tricholemma is no real synapomorphy (cf. above). In the annum genus Avena, diploid Avena longiglumis (2n=14) with the AA genome is basal to sequences of hexaploid Arena sativa (2n=42) with the AACCDD genome (RAJHATHYÆ THOMAS 1974; Fig. 1). Recently, it was shown by genomic in situ hybridisation that it is not possible to discriminate the two genomes A and D, whereas the C genome is different from both other genomes and clearly detectable in metaphase chromosomes (CHEN & ARMSrRON6 1994, JELLEN & al. 1994, LEGGErT & MARKHAND 1995). Probably, the two slightly different, but in neighbor-joining distance method separable repeat types represent the genomes (AD) and C: nos. 1, 2, 6 and 3, 4, 5 (Fig. 1). Three ITS repeats from tetraploid Arrhenatherum elatius (2n--28) are distinguishable from the sequences of all other Avenineae investigated (Fig. 1). Concerning the taxa of Aveneae subtribe Aveninae in a narrow sense (cf. TSVELEV 1976, CONERT 1979--1994), species of Helictotrichon subg. Helictotrichon, H. jahandiezii and Arrhenatherum elatius are the closest relatives of the species of Arena. The subgenera Pratavenastrum and Pubavenastrum occupy a rather remote position. A v e n e a e a n d n e i g h b o r i n g t r i b e s o f s u b f a m . P o o i d e a e . The molecular phylogenetic relationships of tribe Aveneae to other grasses is shown in Fig. 2. This dendrogram contains species of the tribes Bromeae (Bromus inermis), Triticeae (Secale cereale), Poeae (Festuca mairei, Lolium perenne, Dactylis glomerata), and further taxa of Aveneae (incl. Agrostideae sensu MACFARLANE Æ WATSON 1982) that were not included in Fig. 1 (Trisetum, 244 B. GREBENSTEIN84 al.: Koeleria, Briza, Deschampsia, Holcus, Phalaris, Alopecurus). Some data as denoted in Fig. 2 were taken flora HS~AO& al. (1994, 1995b). Helictotrichon subg. Helictotrichon is represented by H. convolutum (no. 1) and H. cantabricum (no. 1), H. subg. Pratavenastrum by H. aetolicum (no. 1) and H. armeniacum (no. 1). The datasets of Arrhenatherum elatius, Avena sativa, Helictotrichon jahandiezii, H. pubescens, Koeleria spp. and Trisetum spp. (cf. Table 2) were reduced to one ITS sequence, respectively. The results in the Aveninae genera Helictotrichon, Avena, and Arrhenatherum demonstrated in Fig. 1 are congruent with this neighbor-joining tree (Fig. 2): Helictotrichon is not resolved as monophyletic. Its four subgenera appear as separate lineages; the larger ortes (subgg. Helictotrichon and Pratavenastrum) are confirmed as monophyletic groups with rather distant relations to each other which is supported by the strongly contrasting distribution pattems of specific satellite DNAs in the genomes of these taxa (GREBENSTEIN & al. 1996). Helietotrichon jahandiezii (subg. Tricholemma) again appears as a sistergroup of subg. Helictotrichon. Subgenus Pratavenastrum is grouping together with Deschampsia cespitosa, a member of an almost cosmopolitan and largely hygrophilous genus with certain morphological affinities to the genus Helictotrichon as adequately addressed by CLAYTON & RENVOIZE (1986). In narrow systematic concepts (e.g. TSVELEV 1976, CONERT 1979--1994), Deschampsia is assigned to a subtribe of Aveneae different from the Aveninae (subtribe Airinae), whereas other monographs treat Deschampsia as part of a larger subtribe Aveninae (PmGER 1949, 1954; CLAYTON & RENVOIZE 1986). Of particular interest are the results for the genera Holcus, Phalaris and Briza, becanse these were not always included in the tribe Aveneae: The ITS sequence of Helietotrichon pubescens (subg. Pubavenastrum) is grouping with that of Alopecurus vaginatus in the same branch of the dendrogram (Fig. 2). In current systematic concepts, the genus Alopecurus is placed together with Phleum and several other genera either in the subtribe Alopecurinae as part of the Aveneae (CLAYTON & RENVOIZE 1986) or in the separate tribe Phleeae (CONERT 1979-1994; TSVELEV 1976, 1989) which, however, is considered to be very close to Aveneae. Together with Holcus lanatus, the species Phalaris truncata and Briza minor are basal to a lineage which consists of 'core' taxa of the Avenineae, i.e. Helictotrichon subg. Helietotrichon, H. subg. Tricholemma, Avena and Arrhenatherum, and, unexpectedly, of the genera Koeleria and Trisetum which are frequently treated together with the genera Trisetara and Rostraria under the separate tribe Koeleriinae. Phalaris was considered either as part of the Aveneae (subtribe Phalaridinae) or, with the same circumscription and genus content, as neighbor tribe of the Aveneae (tribe Anthoxantheae syn. Phalarideae). According to all available information the conspicuous genus Briza was regarded in morphologically-based systematic treatments always as member of the tribe Poeae (PmG~R 1954, POTZTAL 1964, MACFARLANE & WATSON 1982, TSWLEV 1976, CONERT 1979--1994, CLAYTON & R~NVOIZE 1986, MACFARLANE 1987). Briza appeared to be linked with the genus Poa via intermediate types of floret morphology (CLAYTON & RENVOIZE 1986). In contrast, concurrent evidence from restriction site analysis of chloroplast DNA (SORENG & al. 1990) and rDNA ITS data (HslAo & al. 1995b), suggests an alignment of Briza with the Aveneae. This Molecular systematics of Aveneae 245 20 Avena sativa A. longiglumis* 30 40 50 60 GACCA.~~z~CAGACCGAGCACGCGTTATCTATTCCTACTGAGTGGCGGCACCGT-C-GTC .................................. Arrhenatherum eIatius KoeIeria digorica 1 . . . . . . . . . . . . . . G ............. C ..... - . . A . , C . G - . -..C. A...-.-..T ùG . . . . . C . . . . . . . . T . . - . C . . . . . . . . . . . . . . . . . C ..... , .C.GC. ùC.AG.G..-.C.C. . . . . . . . . . . . . . . C . . . . . . . . C. , .C.GC. .C.AG.G..-.C.C. . . . . . . . . . . . . . . C . . . . . . . . C. , .C.GC. ùC.AG.G..-.C.C. . . . . . . . . . . . . . . C ..... , .C.GC. .C.AG.G..-.C.C. . . . . . . . . . . . . . . C..G..-..C. Trisetum flavescens 1 . . . . . . . . . . . . . . C . . . . . . . . T. flavescens 2 T. turcicum I "1". turcicum 2 . . . . . . . . . . . . . . C ..... . . . . . . . . . . . . . . C . . . . . . . . . . . . . . . . . . . . . . . C ..... Helictotrichon cantabricum . . . . . . . . . . . . . . . C . . . . . . . . H. jahandiezii H. aetolicum H. pubescens . . . . . . . . . . . . . . . C .... . . . . . . . . T. A ...... A... . . . . . . . . . . . . . . . T . . . . . . . . C ...... Alopecurus vaginatus . . . . . . . . . . . . . . . C . . . . . . . . . . . . . . A--.. Holcus lanatus Briza minor * Deschampsia cespitosa C . . . . . . . . -... ............ K. K. K. K. digorica 2 digorica 3 pyramidata 1 pyramidata 2 Phalaris truncata * D a c ~ l i s glomerata Festucamairei* Loliumperenne * * .... * 2' . . . . . . - . . . . . . . . . . TT. -..C. ,C.GC. .C.AG.G..-.C.C. ùC.AG.G..-.C.C. C.C.C.GC. -..C.C.C.GC. ùC.AG.G..-.C.C. C...C.GC. ùC.AG.G..-.C.C. -..C...C.GC. C... ùC.AG.G..-.C.C. C. -G.. C... C. GC.. C. . . . . . . . . . . . . . . . T . . . . . . . . C... C. A-... . . . . . . . . . . . . . . . T. A ...... CC .... A... . . . . . . . . T ...... T .... ........ A ...... C ............ C... ....... GC. --.. C.. C ..... C . . . . . . . . . . . . A...A...C.-G..CG.C...C 2' . . . . . . . . . . . . . . . . . . . . . . . . . . . CG. C.. -.. -C. G.. GC. ACT C ....... TGCCG. C . . . . . . . . GG. C.. CGATG. --GCTG... G. C. TAG. CG. -G..CG...G.C C... -. T... CTT. A ..... C.. C. G. C .... . G.C ....... ....... C... .T... -... -. C... -. C.. T -. CA. , -. T... GTC-. T... G. A. T. CAC. T. G.. T . . . . . . . . C ..... -. . . . . . . . . . . TGCCG. C.A-...G. A . . . . . . . . GTC. GC. T...- GT. -... -T-.., T...-.T.C. T...- .C.., Fig. 3. Detail of aligned ITS 1 rDNA sequences of Aveneae and Poeae. Numbers indicate the consecutive bp-positions (5' to 3') from the beginning of ITS1 region, dots denote identity with reference sequence Arena sativa (no. 1), and dashes denote gaps. Sequences of ITS 1 of the genera Koeleria and Trisetum are characterised by a 9-bp deletion at bppositions 45-53. Asterisk designates sequences taken from HSIAO & al. (1994, 1995b) suggestion is fully supported by the broader sampling of Aveneae species and closely allied genera presented here. The genera Koeleria and Trisetum share in the alignment of ITS 1 sequences a 9-bp deletion at bp-positions 45-53 with respect to the reference sequence of Arena sativa (no. 1; Fig. 3). This deletion is present in every sequence of independent clones from different species (Koeleria digorica, K. pyramidata, Trisetum flavescens, T. turcicum). A deletion of these nucleotides from ITS1 evidently occurred prior to the separation of the genera Koeleria and Trisetum, since a convergent event of deletion of the same stretch of nucleotides is not very likely. The consequently suggested monophyly of KoeIeria and Trisetum is further supported by data on highly repetitive DNA sequences (GREBENSTEIN, unpubl.), strong morphological affinities, and the occurrence of intermediate species which makes the taxonomic separation of both genera somewhat arbitrary (cf. CLAYTON & RENVOrZE 1986). The 9-bp deletion in the ITS1 region is a very promising molecular marker to assess phylogenetic relationships in a considerable number of Aveneae genera, especially the ones which on morphological grounds appear to be closely related to Koeleria or Trisetum (annual genera Trisetaria and Rostraria with Mediterranean to Middle East distribution) or show at least certain affinities (annual Old World genera Ventenata, Gaudiniopsis, and Pilgerochloa; New World perennials Graphephorum, Peyritschia, and perennial/annual Sphenopholis). The 246 B. GREBENSTEIN8z al.: presence of the 9-bp deletion in ITS 1 is expected therefore to be characteristic of a particular phylogenetic lineage within Aveneae that is not unambiguously resolved to date by morphologically- and anatomically-based work, and it might contribute to a more precise definition of infratribal taxonomic units of Aveneae. Systematic implications. Our data on the 9-bp deletion in ITS1 sequences supports the taxonomic separation of Trisetum and Helictotrichon. In morphological terms these genera are not strongly delimited (glabrous versus hairy ovaries, hut with exceptions), and the different types of lodicule shapes (Trisetum type, Helictotrichon type, Arena type, etc.) do not represent absolutely reliable diagnostic characters: The 'Trisetum type' occurs not only in Trisetum, but also in Helictotrichon subg. Pubavenastrum, the 'Arena type' in H. subg. Pratavenastrum, and the lodicules of H. jahandiezii are intermediate between the 'Trisetum' and 'Helictotrichon' type (cf. BAUM 1968, RöSER 1989, LANGE 1995). The tribe Aveneae was frequently split up into two different tribes, the Aveneae s. str. containing the genera with many-flowered spikelets (Helictotrichon, Arena, Arrhenatherum, Trisetum, etc.) and the Agrostideae (Alopecurus, Phalaris, Agrostis, Phleum, etc.) with single-flowered spikelets (KUNTH 1815, BENTHAM 1883, HACKEL 1887, HOLUB 1958, HUBBARD 1959, PRAT 1960, PARODI 1961, MACFARLANE Æ WATSOY 1982). Following the arguments of PIL6ER (1949, 1954) recent systematic treatments of the family of grasses usually regarded them as a single tribe: STEBBINS & CRANPTON (1961), TSVELEV (1976), CONERT (19791994), CLAVTON & RENVOIZE (1986), and MACFARLANE(1987) under the correct name 'trib. Aveneae', TATEOKA (1957) under "trib. Agrosteae", KOYAMA(1987) under "trib. Agrostideae". Our data on the ITS sequences (Fig. 2) confirm the latter view with the maintenance of a larger tribe Aveneae. In accordance with current systematic opinions, the tribe Aveneae is closely associated with the tribe Poeae. Both tribes can be regarded as monophyletic sistergroups (Fig. 2) with a common, monophyletic origin in the grass subfam. Pooideae as already indicated by chloroplast DNA restriction site variation (SORENG& al. 1990), by ITS sequence analysis (HsIAO & al. 1994, 1995b), and, though less extensively sampled, by rbcL sequence data (DUVALL & MORTON 1996). A completely different systematic view was expressed in the last systemafic review of grasses published by TSVELEV (1989) who argued that separation of the tribes Poeae and Aveneae is based on comparatively weak morphological characters and consequently summarised them under a broad tribe Poeae. This unconventional suggestion should be kept in mind when further molecular data of phylogenetically critical taxa become available by molecular work. The tribe Triticeae (represented hefe by Secale cereale) and the Bromeae (Bromus inermis) are rather distantly related to Poeae and Aveneae species as indicated by ITS sequence data (Fig. 2; HSIAO 1995b), studies based on chloroplast DNA restriction fragment length polymorphisms (SORENG& al. 1990, KELLOGG 1992), and by morphological evidence (cf. CLAYTON & RENVOIZE 1986). In conclusion, the molecular analysis of ITS sequences of several taxa of the Aveneae suggests that (i) the ancestry of the agronomically important genus Avena has to be sought within comparatively small-flowered Aveneae taxa, (ii) genus Arrhenatherum and small-flowered subgenera of Helictotrichon are close extant Molecular systematics of A v e n e a e 247 relatives of A v e n a , (iii) genus H e l i c t o t r i c h o n is para- if not polyphyletic, (iv) genera T r i s e t u m , K o e l e r i a and probably others form a separate lineage characterised by a particular 9-bp deletion, (v) the delineation of some genera and subtribes of A v e n e a e , and perhaps tribes of subfam. P o o i d e a e needs to be reevaluated by including phylogenetically critical taxa and combining morphological, anatomical and molecular datasets. The support of this work by a grant of the Deutsche Forschungsgemeinschaft within the project: "Molekulare Grundlagen der Evolution bei Pflanzen" is gratefully acknowledged. References BALDWIN, B. G., 1992: Phylogenetic utility of the internal transcribed spacers of the nuclear ribosomal DNA in plants: An example from the Cornpositae. - Molec. Phylogenet. Evol. 1: 3-16. -- SANDERSON, M. J., PORTER, M. J., WOJCIECHOWSKI, M. F., CAMPBELL, C. S., DONOGHUE, M. J., 1995: The ITS region of nuclear ribosomal DNA: A valuäble source of evidence in angiosperm phylogeny. - Ann. Missouri Bot. Gard. 82: 247-277. BAUM, B. R., 1968: Delimitation of the genus A r e n a L. - Canad. J. Bot. 46: 121-132. BENNETT, M. D., 1972: Nuclear DNA content and minimum generation time in herbaceous plants. - Proc. Roy. Soc. London, Ser. B, Biol. Sci. 181: 109-135. 1987: Variation in genomic form in plants and its ecolological implications. - New Phytol. 106 (Suppl.): 177-200. BENTHAM, G., 1883: G r a m i n e a e . - In BENTHAM, G., HOOKER, J. D., (Eds): Genera plantarum, 3 (2), pp. 1074-1215. - London: Reeve. BERNARDI, G., OLOFSSON, B., FILIPSKI, J., ZERIAL, M., SALINAS,J., CUNY, G., MEUNIERROTIVAL, M., RODIER, F., 1985: The mosaic genome of warm blooded vertebrates. Science 228: 953-958. - MOUCHIROUD,D., GAUTIER, C., 1988: Compositional patterns in vertebrate genomes: Conservation and change in evolution. - J. Molec. Evol. 28: 7-18. BUCKLER, E. S. IV, HOLTSFORD,T. P., 1996a: Z e a systematics: ribosomal ITS evidence. Molec. Biol. Evol. 13: 612-622. 1996b: Z e a ribosomal repeat evolution and mutation patterns. - Molec. Biol. Evol. 13: 623-632. CAMPBELL, C. S., WOJCIECHOWSKI, M. F., BALDWIN, B. B., ALICE, L. A., DONOGHUE, M. J., 1997: Persistent nuclear ribosomal DNA sequence polymorphism in the A m e l a n c h i e r agamic comple× ( R o s a c e a e ) . - Molec. Biol. Evol. 14: 81-90. CHEN, Q., ARMSTRONG, K., 1994: Genomic in situ hybridization in A r e n a sativa. Genome 37: 607-612. CLAYTON, W. D., RENVO1ZE, S. A., 1986: Genera graminum. Grasses of the world. - Kew Bu11., Add. Ser. 13.-London: HMSO. CONERT, H. J., 1979-1994: G r a m i n e a e . - In CONERT, H. J., HAMANN, U., SCHULTZEMOTEL, W., WAGENITZ, G., (Eds): G. HEGI, Illustrierte Flora von Mitteleuropa, 3rd edn., 1 (3), pp. 1-560. - Hamburg: Parey; Berlin: Blackwell. CONNOR, H. E., 1979: Breeding systems in the grasses. - New Zealand J. Bot. 17: 547-574. - 1987: Reproductive biology in the grasses. - In SODERSTROM, T. R., HILU, K. W., CAMPBELL, C. S., BARKWORTH,M. E., (Eds): Grass systematics and evolution, pp. 117132. - Washington, D.C.: Smithsonian Institution Press. DUVALL, M. R., MORTON, B. R., 1996: Molecular phylogenetics of P o a c e a e : An expanded analysis of r b c L sequence data. - Molec. Phylogenet. Evol. 5: 352-358. FEDOROV, A. A., 1969: Chromosome numbers of flowering plants. - Leningrad: Nauka. 248 B. GREBENSTEIN 8¢ al.: FELSENSTEIN,J., 1985: Confidence limits of phylogenies: An approach using the bootstrap. - Evolution 39: 783-791. FREY, L., 1991: Taxonomy, karyology and distfibution of selected genera of tribe A v e n e a e ( P o a c e a e ) in Poland: I. A v e n u l a . - Fragm. Florist. Geobot. 35: 101-137. GOLDBLATT, P., JOHNSON, D. E., 1981-1996: Index to plant chromosome numbers. Monogr. Syst. Bot. Missouri Bot. Gard. 5, 8, 13, 23, 30, 40, 51, 58. - St. Louis, Missouri: Missoufi Botanical Garden. GREBENSTEIN, B., GREBENSTEIN, O., SAUER, W., HEMLEBEN, V., 1995: Characterization of a highly repeated DNA component of perennial oats ( H e l i c t o t r i c h o n , P o a c e a e ) with sequence similarity to a A-genome specific satellite DNA of rice ( O r y z a ) . - Theor. Appl. Genet. 90: 1101-1105. .... 1996: Distfibution and complex organisation of satellite DNA sequences in A v e n e a e species. - Genome 39: 1045-1050. HACKEL, E., 1887: G r a m i n e a e . - In ENGLER, A., 13RANTL, K., (Eds): Die natürlichen Pflanzenfamilien nebst ihren Gattungen und wichtigeren Arten insbesondere den Nutzpflänzen, 2 (2), pp. 1-97. - Leipzig: Engelmann. HAMBY, R. K., ZIMMER, E. A., 1988: Ribosornal RNA sequences for inferring phylogeny within the grass family ( P o a c e a e ) . - Pl. Syst. Evol. 160: 29-37. HEMLEBEN, V., 1993: Repetitive and highly repetitive DNA components as molecular markers for evolutionary studies and in plant breeding. - Current Topics Molec. Genet. (Life Sci. Adv.) 1: 173-185. HOLUB, J., 1958: Bemerkungen zur Taxonomie der Gattung H e l i c t o t r i c h o n BESS. - In KLÄSTERSKX~, I., (Ed.): PHILIPP MAXMILIAN OPIZ und seine Bedeutung für die Pflanzentaxonomie, pp. 101-133. - Praha: Tschechoslowakische Akademie der Wissenschaften. - 1962: Ein Beitrag zur Abgrenzung der Gattungen in der Tribus A v e n e a e : Die Gattung A v e n o c h l o a HOLUB. - Acta Horti Prag. Bot. 1962: 75-86. - 1976: A newly adopted restriction of illegitimäcy in generic names and its consequence for A v e n o c h l o a HOLUB. --Folia Geobot. Phytotax. 11: 281-300. HSIAO, C., CHATTERTON, N. J., ASAY, K. H., JENSEN, K. B., 1994: Phylogenefic relationships of 10 grass species: An assessment of phylogenetic utility of the internal transcribed spacer region in nuclear ribosomal DNA in monocots. - Genome 37:112120. .... 1995a: Phylogenetic relationships of the monogenomic species of the wheat tribe, T r i t i c e a e ( P o a c e a e ) , inferred flora nuclear rDNA (internal transcribed spacer) sequences. - Genome 38: 211-223. .... 1995b: Molecular phylogeny of the P o o i d e a e ( P o a c e a e ) based on nuclear rDNA (ITS) sequences. - Theor. Appl. Genet. 90: 389-398. HUBBARD, C. E., 1959: G r a m i n e a e . - In HUTCHINSON, J., (Ed.): The families of flowering plants, 2nd edn. 2. Monocotyledons, pp. 710-741. - Oxford: Clarendon Press. JELLEN, E. N., GILL, B. S., Cox, T. S., 1994: Genomic in situ hybridisation differentiates between A/D- and C-genome chromatin and detects intergenomic translocafions in polyploid oat species (genus A v e n a ) . - Genome 37: 613-618. JOBST, J., RING, K., HEMLEBEN,V., 1998: Molecular evolution of the internal transcribed spacer (ITS1 and ITS2) and phylogenetic relationships among species of the family C u c u r b i t a c e a e . - Molec. Phylogenet. Evol. 9: 204-219. JUKES, T. H., CANTOR, C. R., 1969: Evolution of protein molecules. - In MURNO, H. N., (Ed.): Mammalian protein metabolism, pp. 21-132. - New York: Academic Press. KELLOGG, E. A., 1992: Tools for studying the chloroplast genome in the T r i t i c e a e ( G r a m i n e a e ) : An EcoRI map, a diagnostic deletion, and support for B r o m u s as an outgroup. - Amer. J. Bot. 79: 186-197. Molecular systematics of Aveneae 249 KERGUiLEN, M., 1975: Les Gramineae (Poaceae) de la flore fran~aise. -Lejeunia, n. s. 75: 1-343. KIM, K. J., JANSEN, R. K., 1994: Comparison of phylogenetic hypothesis among different data sets in dwarf dandelions (Krigia). Additional information from internal transcribed spacer sequences of nuclear ribosomal DNA. - Pl. Syst. Evol. 177: 53-69. KOYAMA, T., 1987: Grasses of Japan and its neighboring regions. - Tokyo: Kodansha. KVNTH, C. S., 1815: Considérations générales sur les graminées. - Mém. Mus. Hist. Nat. 2: 62-75. LANGE, D., 1995: Untersuchungen zur Systematik und Taxonomie der Gattung HeIictotrichon Besser ex J. A. SCHULTES & J. H. SCHULTES (Poaceae) in Südosteuropa und Vorderasien. - Biblioth. Bot. 144: 1-238. LEGGETT, J. M., MARKHAND, G. S., 1995: The genomic structure of Arena revealed by GISH. - In BRANDHAM,P. E., BENNETT, M. D., (Eds): Kew Chromosome Conference IV, pp. 133-139. - Richmond: Royal Botanic Gardens, Kew. MACEARLANE, T. D., 1987: Poaceae subfamily Pooideae. - In SODERSlgOM, T. R., HILU, K. W., CAMPBELL, C. S., BARKWORTH, M. E., (Eds): Grass systematics and evolution, pp. 265-276. - Washington, D.C.: Smithsonian Institution Press. - WATSON, L., 1982: The circumscription of Poaceae subfamily Pooideae, with notes on some controversial genera. - Taxon 29: 645-666. MAGGINI, F., BALDASSINI, S., 1995: Ribosomal RNA genes in the genus Pinus. Caryologia 48: 17-25. M¥ERS, E. W., MILLER, W., 1988: Optimal alignment in linear space. - Computer Applic. Biosci. 4: 11-17. PARODI, L. R., 196l: La taxonomfa de las Gramineae argentinas a la luz de las investigaciones mäs recientes. - Recent Adv. Bot. 1: 125-130. PILGER, R., 1949: Additamenta agrostologica. - Bot. Jahrb. Syst. 74: 1-27. 1954: Das System der Gramineae unter Ausschluß der Bambusoideae. - Bot. Jahrb. Syst. 76: 281-384. POTZTAL, E., 1964: 7. Reihe Graminales (Poales, Glumiflorae p.p.). - In MELCHIOR, H., (Ed.): A. ENGLERS Syllabus der Pflanzenfamilien, 2nd edn. 2, pp. 561-579. - Berlin: Borntraeger. PRAT, H., 1960: Vers une classification naturelle des graminées. - Bull. Soc. Bot. France 107: 32-79. PROBATOVA, N. S., SOKOLOVSKAYA,A. P., 1978: Chromosome numbers and taxonomy of some grasses in Caucasus. - Bot. Zhurn. (Moscow & Leningrad) 63: 1121-1131. [In Russian.] RAJHATHY, T., THOMAS, H., 1974: Cytogenetics of oats (Avena L.). - Misc. Publ. Geriet. Soc. Canada 2: 1-90. ROMERO ZARCO, C., 1984: Revisión taxonómica del género Avenula (DuMORT.) DUMORT. en la Penfnsula Ibérica e Islas Baleares. - Lagascalia 13: 39-146. 1985: Revisión del género Arrhenatherum RouY en la Penfnsula Ibérica. - Acta Bot. Malac. 10: 123-154. RÖSER, M., 1989: Karyologische, systematische und chorologische Untersuchungen an der Gattung Helictotrichon BESSER ex SCHULTES • SCHULTES im westlichen Mittelmeergebiet. - Diss. Bot. 145: 1-250. 1995: Typification of the names of two widespread Eurasian grass species: Arena pratensis L. and A. pubescens HUDS. (Poaceae." Helictotrichon). - Taxon 44: 395-399. - 1996: Ecogeography of the grass genus Helictotrichon (Pooideae: Aveneae) in the Mediterranean and adjacent r e g i o n s . - Pl. Syst. Evol. 203: 181-281. - 1997: Patterns of diversification in Mediterranean oat grasses (Poaceae: Aveneae). Lagascalia 19: 101-120. - - - 250 B. GREBENSTEIN Æ al.: Molecular systematics of A v e n e a e SALINAS, J., MATASSL G., MONTERO, L. M., BERNARDI, G., 1988: Compositional compartmentalization and compositional patterns in the nuclear genomes of plants. Nucl. Acids Res. 16: 4269-4285. SORENG, R. J., DAVIS, J. I., DOYLE, J. J., 1990: A phylogenetic anälysis of chloroplast DNA restriction site variation in P o a c e a e subfam. P o o i d e a e . - Pl. Syst. Evol. 172: 83-97. STEBBINS, G. L., CRAMPTON, B., 1961: A suggested revision of the grass genera of temperate North America. - Recent Adv. Bot. 1: 133-145. TATEOKA,T., 1957: Miscellaneous papers on the phylogeny of P o a c e a e (10). Proposition of a new phylogenetic system of P o a c e a e . - J. Jap. Bot. 32: 275-287. TORRES, R. A., GANAL, M., HEMLEBEN,V., 1990: GC balance in the internal transcribed spacers ITS1 and ITS2 of nuclear ribosomal RNA genes. - J. Molec. Evol. 30: 170-181. TSVELEV, N. N., 1976: Zlaki SSSR. - Leningrad: Nauka. [In Russian.] 1989: The system of grasses ( P o a c e a e ) and their evolution. - Bot. Rev. 55: 141-204. (Original published 1987 in Russian.) VAN DE PEER, Y., DE WACHTER, R., 1993: Treecon: a software package for the construction and drawing of evolutionary trees. - Computer Applic. Biosci. 9:177-182. VIERHAPPER,F., 1914: Zur Systematik der Gattung A r e n a . - Verh. Ges. Deutsch. Naturf. Ärzte 85: 670-674. - Addresses of the authors: B. GREBENSTEIN,W. SAUER,Lehrstuhl für Spezielle Botanik, Botanisches Institut, Universität Tübingen, Auf der Morgenstelle 1, D-72076 Tübingen, Germany. - V. HEMLEBEN (correspondence), Lehrstuhl für Allgemeine Genetik, Biologisches Institut, Universität Tübingen, Auf der Morgenstelle 28, D-72076 Tübingen, Germany. - M. RöSER, Institut für Botanik, Universität Leipzig, Johannisallee 21-23, D-04103 Leipzig, Germany. Accepted August 16, 1997 by F. EHRENDORFER