1094

Genome size variation in Hordeum spontaneum populations Timo Turpeinen, Johanna Kulmala, and Eviatar Nevo

Abstract: Populations of wild barley, Hordeum spontaneum (C. Koch), originating from 10 ecologically and geographically different sites in Israel, were assessed for genome size. Measurements were obtained by flow cytometry using propidium iodide staining. Genome sizes ranged from 9.35 to 9.81 pg. Variance analysis indicated a significant difference between populations. Genome sizes were positively correlated with mean January temperature. Our results corroborate previous findings of intraspecific variation in genome size from different plant species. The positive correlations between climate and genome size suggest that the latter is adaptive and determined by natural selection. Key words: Hordeum spontaneum, genome size, flow cytometry, intraspecific variation, natural selection. Résumé : La taille du génome a été évaluée chez des populations d’orge sauvage, Hordeum spontaneum (C. Koch), provenant de 10 sites distincts sur les plans écologique et géographique en Israël. Les mesures ont été réalisées par coloration à l’iodure de propidium suivie de cytométrie en flux. La taille du génome variait entre 9,35 et 9,81 pg. Une analyse de variance a indiqué qu’il y avait des différences significatives entre populations. Une corrélation positive entre la taille du génome et la température moyenne au mois de janvier a été observée. Ces résultats corroborent des résultats antérieurs qui avaient montré de la variation intraspécifique quant à la taille du génome chez d’autres espèces. La corrélation positive entre le climat et la taille du génome suggère que ce dernier caractère est adaptatif et qu’il est déterminé par la sélection naturelle. Mots clés : Hordeum spontaneum, taille du génome, cytométrie en flux, variation intraspécifique, sélection naturelle. [Traduit par la Rédaction]

Introduction

1099

Turpeinen et al. has become an important germplasm source for introducing

Barley, Hordeum vulgare L., is one of the principal cereal crops in the world crop and is cultivated in all temperate areas (von Bothmer et al. 1995). Wild barley, Hordeum spontaneum, has been considered to be a progenitor of cultivated barley and is distributed throughout the Near East from western Turkey eastward into Southwest Asia (Harlan and Zohary 1966). Wild barley represents a rich genepool, and several studies concerning evolution, adaptation, and population genetics of wild barley have been conducted (Nevo et al. 1979, 1986; Nevo 1992). Hordeum spontaneum has the same chromosome number (2n = 14) as H. vulgare, and there is no biological crossing barrier between the two species. Modern plant breeding has narrowed the genetic base of cultivated barley (Nevo et al. 1986), and wild barley Corresponding Editor: P.B. Moens. Received January 4, 1999. Accepted June 3, 1999. T. Turpeinen.1 Agricultural Research Centre of Finland, Plant Production Research, Crops and Soil, FIN-31600 Jokioinen, Finland. J. Kulmala. Department of Biology, University of Joensuu, P.O. Box 111, FIN-80101 Joensuu, Finland. E. Nevo. Institute of Evolution, University of Haifa, Mt. Carmel, Haifa 31905, Israel. 1

Author to whom all correspondence should be addressed (e-mail: [email protected]).

Genome 42: 1094–1099 (1999)

new genes into cultivated barley. Genome sizes have been determined for numerous angiosperm species (http://www.rbgkew.org.uk/cval/database1.html) and genome size variation has been found both between and within species (Bennett and Leitch 1995, 1997). Differences between species are known to vary over 600-fold, ranging from less than 0.2 pg in Arabidopsis thaliana L. to 127.4 pg in Fritillaria assyriaca Baker (Bennett and Leitch 1997), perennials having larger genomes than annuals (Bennett 1972). Intraspecific variation in angiosperm genome size has been reported (Price et al. 1981a, 1981b; Rayburn et al. 1985; Bennett and Bennett 1992; Ceccarelli et al. 1992; Buitendijk et al. 1997; Caceres et al. 1998; Chung et al. 1998) in addition to variation within a single plant (Michaelson et al. 1991). Though there is some disagreement about intraspecific variation, as reviewed by Bennett and Leitch (1997), such observations led to the concept of a plastic plant genome and a C-value. Genome size of flax (Linum usitatissimum L.) was shown to change under certain environmental treatments (Cullis 1983), and the genome size of Microseris bigelovii (Gray) Sch. Bip. has been influenced by selection for low DNA content in a stressful and (or) time-limited environment (Price et al. 1981a). Variation in genome size also has been linked to changes in repetitive DNA amount (Flavell 1986). In the present work, the nuclear DNA content was determined from 97 accessions of 10 H. spontaneum populations by flow cytometry to indicate the existence of inter© 1999 NRC Canada

Turpeinen et al. Fig. 1. Geographic distribution of the 10 populations of Hordeum spontaneum from Israel (modified from Nevo et al. 1979). 1, Mt. Hermon; 9, Mt. Meron; 10, Maalot; 11, Damon; 14, Talpiyyot; 18, Revivim; 20, Sede Boqer; 22, Mehola; 23, Wadi Quilt; 33, Avedat.

1095 Fig. 2. Flow cytometry of Hordeum spontaneum leaf nuclei. The data are displayed as a histogram of numbers of nuclei per relative fluorescence channel (n = 1024) on a linear scale. Peak A, cell debris; peak B, RTRBC nuclei; peak C, the G1-phase leaf nuclei; and peak D, the G2-phase leaf nuclei. A total of 2595 nuclei were counted.

at –20°C in storage buffer (5% DMSO, 250 mM saccharose, 40 mM sodium citrate, pH 7.6) before use. Validity of the frozen RTRBC was checked every week during the experiment. Fluorescence was measured on a linear scale over 1024 channels. A DNA content of 2C = 2.33 pg for chicken red blood cell (CRBC) nuclei (Galbraith et al. 1983) was used to determine the DNA content of RTRBC (2C = 4.97 ± 0.02 pg). The 2C value of the plant nuclear DNA amount was calculated as (channel number of G1 plant nuclear peak/channel number of RTRBC nuclear peak) × 4.97 pg. Events per sample ranged from 1000 to 5000. Analysis of variance (GLM) with Tukey’s studentized range (HSD) test as well as Spearman’s rank correlation, and stepwise multiple regression analysis were performed using an SAS statistical package (SAS Institute Inc. 1989–1996).

population differences and the probable influence of natural selection by environmental stress.

Materials and methods Seeds from 97 accessions of 10 populations (Fig. 1) of H. spontaneum were collected from different sites in Israel, as detailed in Nevo et al. (1979). Seeds were germinated on a Petri dish containing 0.01% CaSO4, 0.8% agar for 1–2 weeks, and then sown in pots in a greenhouse with a 16 h 22°C day : 8 h 18°C night period. For flow cytometry analysis (see Galbraith 1989) 50 mg of leaf material of seedlings was used. Intact nuclei were prepared as described previously (Arumuganathan and Earle 1991). Three separate measurements for each accession were done. A FACSort flow cytometer equipped with a MacIntosh computer system and the CELLQUEST software package (Becton Dickinson, San Jose, Calif.) was used to measure and analyse relative fluorescence intensities from propidium iodide (PI)stained nuclei. An argon–ion laser beam operating at a wavelength of 488 nm was used for excitation. Nuclear DNA content was estimated by the fluorescence of the nuclei with PI relative to internal standard nuclei of rainbow trout red blood cells (RTRBC). Fresh rainbow trout blood was collected into heparinized tubes and kept

Results A typical flow cytometry plot is shown in Fig. 2., where the histogram represents the PI-fluorescence intensity over the 1024 channels. Peak A at the lowest channels corresponds with debris, peak B for RTRBC (G1-phase), peaks C and D for sample (G1- and G2-phase). The coefficient of variation (CV) for samples ranged from 1.16% to 3.07% (mean 1.93, standard deviation 0.35). Variance analyses indicated significant differences between populations for genome size (F = 10.31, P < 0.0001). The mean 2C-values for populations are listed in Table 1, and populations are grouped with Tukey’s studentized range (HSD) test, indicating significant differences between populations at α = 0.05. No statistically significant differences were established within populations. Several ecogeographical factors (Table 2, detailed information in Nevo et al. 1979) linked to growing sites of populations were available for correlation analyses. The Spearman rank correlation test, in addition to stepwise multiple regression, was used to determine associations between genome size and ecogeographical factors. With the Spearman rank correlation test, positive correlation (P = 0.0093) © 1999 NRC Canada

1096

Genome Vol. 42, 1999

Table 1. Nuclear DNA content, standard deviation, and Tukey’s grouping for the 10 populations of Hordeum spontaneum from Israel. Population

Mean 2C-value (pg)

Standard deviation

Tukey’s groupinga

1 9 10 11 14 18 20 22 23 33

9.35 9.48 9.53 9.74 9.81 9.62 9.63 9.69 9.74 9.55

0.18 0.14 0.19 0.23 0.24 0.26 0.27 0.27 0.24 0.31

E D C A A A A A A B

E DE B B B B B C

CD CD C D

Means with the same letter are not significantly different at the α = 0.05 level. a

was found between mean January temperature (Tj), the midgrowth period of H. spontaneum, and mean genome sizes of 10 populations (Table 3). Stepwise multiple regression analysis established also the statistical significance of variable Tj (r = 0.83, P < 0.003, Fig. 3).

Discussion Genome size measurements were done for 10 populations of H. spontaneum representing wide ecological and geographical differences in Israel from the mesic north to the xeric south. The highest genome size value was for the Talpiyyot population no. 14, near the Judean desert (9.81 pg), and the smallest for the high elevation Mount Hermon population no. 1 (9.35 pg). The largest genome contained approximately 5% more DNA than the smallest. Our results showed significant differences between H. spontaneum populations in genome size, which in turn had significant positive correlation with mean January temperature. The importance of January temperature lies in the fact that it is the mid-growing season temperature of wild barley in Israel. The reason for correlation might be that in the conditions where time is a limiting factor for species to reproduce the smaller genome size allows a shorter period of growing and a fast completion of their life cycle as discussed below. Previous measurements for genome sizes in H. spontaneum were made by Bennett and Smith (1971) and Kankaanpää et al. (1996). Our genome size measurements gave lower values than those of Bennett and Smith (Feulgen method) but higher (except one) than those of Kankaanpää et al. (flow cytometry method). These deviations can be the result of different methods (Feulgen vs. flow cytometry) used or, compared with Kankaanpää et al., different staining dyes (DAPI, 4′,6-diamidino-2-phenylindole vs. propidium iodide (PI)) and machines (Partec PAS II vs. FACSort). Dolezel et al. (1992) found a statistically significant difference between DNA contents measured with PI and DAPI. Kankaanpää et al. (1996) had, for example, two accessions (Revivim 18-27 and Sede Boqer 20-48) from the same populations we analysed. Their mean genome size for 18-27 was 9.35 pg compared with ours at 9.62 pg for the mean of pop-

ulation 18 (2.8% difference) and 8.27 pg for 20-48 compared with ours at 9.63 pg for population 20 (16% difference). Their observed pattern of genome size variation in H. spontaneum was not correlated with environmental factors. Genome size measurements seem to give internally consistent value with one method, but they might not be directly comparable with values measured with other methods (A. Schulman, personal communication). Intraspecific variation in genome size has been correlated with environmental factors like altitude in Zea mays L. ssp. mays (Rayburn and Auger 1990) and in Dasypyrum villosum L. (Caceres et al. 1998), latitude in Festuca arundinacea Schreber (Ceccarelli et al. 1992), and evaporation and soil type in H. spontaneum (Vicient et al. 1999). In Microseris douglasii (D.C.) Sch. Bip (Price et al. 1981b) plants with larger genome sizes were generally restricted to high rainfall and well-developed soil (Price et al. 1981b). Interestingly they showed temporal correlation between precipitation and DNA amount in single population, M, near Jolon, Calif., where plants collected in 1973 had approximately 9% higher DNA amount than plants collected in 1962 and 1977, which had in the preceding growing seasons suffered severe drought. They speculated that the selection pressure may be related to the duration of favorable moisture availability in the soil and hence the time available for the completion of the life cycle. Similarly, Price et al. (1981a) suggested in M. bigelovii that plants for lower DNA amounts resulted from selection in stressful and (or) time-limited environments. Bennett and Bennett (1992) have also suggested, at least in part, a similar explanation for lower DNA amount in natural populations of M. effusum compared with plants found in botanical gardens. In F. arundinacea, genome sizes correlated positively with the mean temperature during the year and with that of the coldest month (Ceccarelli et al. 1992). Variation in DNA content within one and the same progeny of Helianthus annuus L. was 14.7% higher in seeds (achenes) collected from the periphery than in seeds collected from the middle of the flowering heads of plants belonging to a line selfed for 10 years (Natali et al. 1993). Variations were based on the redundancy of the repetitive DNA. Their hypothesis to DNA changes was microenvironmental variations, possibly due to differences in the availability of water and nutrients in different portions of the head and (or) to differences in the times at which reproductive events take place, since they occur within the head in centripetal succession. Also, a positive correlation was observed between genome size and flowering time in mature plants from seeds collected from one and the same head. These foregoing observations together with ours are in accordance with nucleotype theory (Bennett 1972) which has been proposed to explain variations in genome sizes in annual plants. Following this theory, genome size is positively correlated to nuclear volume (Baetcke et al. 1967), cell volume (Price et al. 1973; Cavalier-Smith 1978), mitotic cycle time (Van’t Hof and Sparrow 1963; Van’t Hof 1965; Evans and Rees 1971), and the duration of meiosis (Bennett 1971). These characters affect the phenotype by the physical effects on its mass and volume independent of the genome-encoded informational content (Bennett 1972). Bennett (1987) has also pointed out that a causal relationship involving two steps has been established linking DNA amount and any fac© 1999 NRC Canada

Turpeinen et al.

Table 2. Ecogeographical data from the growing sites of 10 populations of Hordeum spontaneum from Israel. Pop.

Site of origin

Long.

Lat.

Alt. (m)

Tm (°C)

Ta (°C)

Tj (°C)

Td (°C)

Tdd (°C)

Rn (mm)

Rd

Hu14

Huan

Dwsm

Ev (mm)

So

1 9 10 11 14 18 20 22 23 33

Mt. Hermon Mt. Meron Maalot Damon Talpiyyot Revivim Sede Boqer Mehola Wadi Quilt Avedat

35.75 35.4 35.27 35 35.23 34.75 34.78 35.48 35.38 34.77

33.28 33.05 33 32.73 31.75 31.02 30.87 32.13 31.83 30.82

1530 1150 500 425 800 320 450 –150 50 525

11 14 17 19 18 20 19 22 23 19

20 22 23 24 24 27 26 30 30 25

1 6 8 11 9 10 9 13 14 9

19 16 15 13 15 15 15 17 16 16

6 8 10 8 9 14 13 13 13 13

1400 1010 785 686 486 130 91 270 170 100

70 65 55 56 42 18 15 39 32 15

52 49 50 59 50 38 36 34 40 36

58 61 64 69 61 55 53 53 55 53

60 50 55 80 40 65 70 22 25 70

160 155 150 140 160 170 168 180 180 168

1 1 2 1 2 8 11 8 8 11

Note: Symbols of variables: Pop., population; Long., longitude (in decimals); Lat., latitude (in decimals); Alt., altitude; Tm, mean annual temperature; Ta, mean August temperature; Tj, mean January temperature; Td, seasonal temperature; Tdd, daily temperature difference; Rn, annual rainfall; Rd, number of rainy days; Hu14, humidity at 14:00; Huan, annual humidity; Dwsm, number of dew nights in summer; Ev, mean annual evaporation. So, soil types: 1, Terra Rossa; 2, Rendzina; 8, Alluvium; 11, Loess.

Table 3. Spearman rank correlations (upper) and probabilities (lower) between ecogeographical factors and mean genome sizes from 10 Hordeum spontaneum populations from Israel.

Genome size Genome size

Long.

Lat.

Alt. (m)

Tm (°C)

Ta (°C)

Tj (°C)

Td (°C)

Tdd (°C)

Rn (mm)

Rd

Hu14

Huan

Ev (mm)

So

–0.25532 0.4765

–0.43161 0.2129

–0.55319 0.0972

0.62156 0.0551

0.58716 0.0743

0.76926 0.0093

–0.45118 0.1906

0.26428 0.4606

–0.40122 0.2505

–0.32317 0.3624

–0.05505 0.8800

0.00310 0.9932

0.24234 0.4999

0.21952 0.5423

Note: For symbols of variables, see Table 2.

1097

© 1999 NRC Canada

1098 Fig. 3. Correlation between the mean January temperature (Tj) and the mean value of genome sizes of 10 populations of Hordeum spontaneum obtained by multiple regression analysis.

tor (such as temperature or water availability) that determines a time-limited environment. Firstly, a causal link has been established between variation in DNA C-value and variation in its temporal phenotypic consequences, such as the minimum time to produce mature seed. Secondly, such temporal consequences can determine the ability of a species to survive and reproduce in particular time-limited environments, so that minimum generation time can be of critical adaptive significance and the phenotypic character on which selection operates. This could also explain, at least in part, our observation of variation in genome size found in H. spontaneum populations. The highest genome size was for population 14 (Talpiyyot) and it was an exception from the observed trend. Its high genome size may indeed be related to its living very near the desert and on poor rendzina soil. We need additional comparative intraspesific studies in nature as well as experimental stress studies to understand more about the role of genome size variation and its correlation to environmental factors.

Acknowledgements We thank Dr. Alan Schulman (University of Helsinki) and two anonymous referees for constructive comments on an earlier version of the manuscript. Dr. Jonathan Robinson (Agricultural Research Centre, Finland) is thanked for his statistical help and for revising the language. Dr. Veli-Matti Rokka, Airi Tauriainen, and Teija Tenhola from the Agricultural Research Centre of Finland are gratefully thanked for their help with flow cytometry.

References Arumuganathan, K., and Earle, E.D. 1991. Estimation of nuclear DNA content of plants by flow cytometry. Plant Mol. Biol. Rep. 9: 229–241. Baetcke, K.P., Sparrow, A.H., Nauman, C.H., and Schwemmer, S.S. 1967. The relationship of DNA content to nuclear and chro-

Genome Vol. 42, 1999 mosome volumes and to radiosensitivity. Proc. Natl. Acad. Sci. U.S.A. 58: 533–540. Bennett, M.D. 1971. The duration of meiosis. Proc. R. Soc. Lond. Ser. B Biol. Sci. 178: 277–299. Bennett, M.D. 1972. Nuclear DNA content and minimum generation time in herbaceous plants. Proc. R. Soc. Lond. Ser. B. Biol. Sci. B 181: 109–135. Bennett, M.D. 1987. Variation in genomic form in plants and its ecological implications. New Phytol. 106: 177–200. Bennett, M.D., and Leitch, I.J. 1995. Nuclear DNA amounts in angiosperms. Ann. Bot. 76: 113–176. Bennett, M.D., and Leitch, I.J. 1997. Nuclear DNA amounts in angiosperms-583 new estimates. Ann. Bot. 80: 169–196. Bennett, M.D., and Smith, J.B. 1971. The 4C nuclear DNA content of several Hordeum genotypes. Can. J. Genet. Cytol. 13: 607– 611. Bennett, S.T., and Bennett, M.D. 1992. Variation in nuclear DNA amount between wild and cultivated populations of Milium effusum (2n = 28). Genome, 35: 1050–1053. Buitendijk, J.H., Boon, E.J., and Ramanna, M.S. 1997. Nuclear DNA content in twelve species of Alstroemeria L. and some of their hybrids. Ann. Bot. 79: 343–353. Caceres, M.E., Pace, C. De., Scarascia Mugnozza, G.T., Kotsonis, P., Ceccarelli, M., and Cionini, P.G. 1998. Genome size variations within Dasypyrum villosum: Correlations with chromosomal traits, environmental factors and plant phenotypic characteristics and behaviour in reproduction. Theor. Appl. Genet. 96: 559–567. Cavalier-Smith, T. 1978. Nuclear volume control by nucleoskeletal DNA, selection for cell volume and cell growth rate, and the solution of the DNA C-value paradox. J. Cell Sci. 34: 247–278. Ceccarelli, M., Falistocco, E., and Cionini, P.G. 1992. Variation of genome size and organization within hexaploid Festuca arundinacea. Theor. Appl. Genet. 83: 273–278. Chung, J., Lee, J.H., Arumuganathan, K., Graef, G.L., and Specht, J.E. 1998. Relationship between nuclear DNA content and seed and leaf size in soybean. Theor. Appl. Genet. 96: 1064–1068. Cullis, C.A. 1983. Environmentally induced DNA changes in plants. CRC Crit. Rev. Plant Sci. 1: 117–131. Dolezel, J., Sgorbati, S., and Lucretti, S. 1992. Comparison of three DNA fluorochromes for flow cytometric estimation of nuclear DNA content in plants. Physiol. Plant. 85: 625–631. Evans, G.M., and Rees, H. 1971. Mitotic cycles in dicotyledons and monocotyledons. Nature (London), 233: 350–351. Flavell, R.B. 1986. Repetitive DNA and chromosome evolution in plants. Philos. Trans. R. Soc. Lond. B Biol. Sci. 312: 227–242. Galbraith, D.W. 1989. Analysis of higher plants by flow cytometry and cell sorting. Int. Rev. Cytol. 116: 165–228. Galbraith, D.W., Harkins, K.R., Maddox, J.M., Ayres, N.M., Sharma, D.P., and Firoozabady, E. 1983. Rapid flow cytometry analysis of the cell cycle in intact plant tissues. Science (Washington D.C.), 220: 1049–1051. Harlan, J.R., and Zohary, D. 1966. Distribution of wild wheats and barley. Science (Washington D.C.), 153: 1074–1080. Kankaanpää, J., Mannonen, L., and Schulman, A.H. 1996. The genome sizes of Hordeum species show considerable variation. Genome, 39: 730–735. Michaelson, M.J., Price, H.J., Johnston, J.S., and Ellison, J.R. 1991. Variation of nuclear DNA content in Helianthus annuus (Asteraceae). Am. J. Bot. 78: 1238–1243. Natali, L., Cavallini, A., Cionini, G., Sassoli, O., Cionini, P.G., and Durante, M. 1993. Nuclear DNA changes within Helianthus © 1999 NRC Canada

Turpeinen et al. annuus L.: Changes within single progenies and their relationships with plant development. Theor. Appl. Genet. 85: 506–512. Nevo, E. 1992. Origin, evolution, population genetics and resources for breeding of wild barley, Hordeum spontaneum, in the fertile crescent. In Barley: Genetics, biochemistry, molecular biology and biotechnology. Edited by P.R. Shewry. C.A.B. International, Wallingford, U.K. pp. 19–43. Nevo, E., Zohary, D., Brown, A.H.D., and Haber, M. 1979. Genetic diversity and environmental associations of wild barley, Hordeum spontaneum, in Israel. Evolution, 33: 815–833. Nevo, E., Beiles, A., and Zohary, D. 1986. Genetic resources of wild barley in the Near East: structure, evolution and application in breeding. Biol. J. Linn. Soc. 27: 355–380. Price, H.J., Sparrow, A.H., and Nauman, A.F. 1973. Correlations between nuclear volume, cell volume and DNA content in meristematic cells of herbaceous angiosperms. Experientia, 29: 1028–1029. Price, H.J., Chambers, K.L., and Bachmann, K. 1981a. Genome size variation in diploid Microseris bigelovii (Asteraceae). Bot. Gaz. 142: 156–159. Price, H.J., Chambers, K.L., and Bachmann, K. 1981b. Geographic and ecological distribution of genomic DNA content variation in Microseris douglasii (Asteraceae). Bot. Gaz. 142: 415–426.

1099 Rayburn, A.L., and Auger, J.A. 1990. Genome size variation in Zea mays ssp. mays adapted to different altitudes. Theor. Appl. Genet. 79: 470–474. Rayburn, A.L., Price, H.J., Smith, J.D., and Gold, J.R. 1985. Cband heterochromatin and DNA content in Zea mays. Am. J. Bot. 72: 1610–1617. SAS Institute Inc. 1989–1996. SAS Software package version 6.12. SAS Institute Inc., Cary, N.C. Van’t Hof, J. 1965. Relationships between mitotic cycle duration, S period duration and the average rate of DNA synthesis in the root meristem of several plants. Exp. Cell Res. 39: 48–58. Van’t Hof, J., and Sparrow, A.H. 1963. A relationship between DNA content, nuclear volume, and minimum mitotic cycle time. Proc. Natl. Acad. Sci. U.S.A. 49: 897–902. Vicient, C.M., Suoniemi, A., Anamthawat-Jónsson, K., Tanskanen, J., Beharav, A., Nevo, E., and Schulman, A.H. 1999. Retrotransposon BARE-1 and its role in genome evolution in the genus Hordeum. Plant Cell, 11: 1769–1784. von Bothmer, R., Jacobsen, N., Baden, C., Jørgensen, R.B., and Linde-Laursen, I. 1995. An ecogeographical study of the genus Hordeum. International Plant Genetic Resources, Rome. p. 29.

© 1999 NRC Canada