J Appl Genet 49(3), 2008, pp. 213–220

Original article

Genetic diversity of ornamental and cultivars assessed with isozymes

Agnieszka Krzymiñska1, Magdalena Gaw³owska2, Bogdan Wolko2, Jan Bocianowski3

1Department of Ornamental , University of Life Sciences, Poznañ, Poland 2Institute of Genetics, Polish Academy of Sciences, Poznañ, Poland 3Department of Mathematical and Statistical Methods, University of Life Sciences, Poznañ, Poland

Abstract. The aim of the study was to verify the similarity of 13 species and 5 cultivars of ornamental and classify them into groups based on morphological and isozyme variation. The work embraced: , A. caeruleum, A. christophii, A. giganteum, A. karataviense, A. moly, A. nigrum, A. pyrenaicum, A. rosenbachianum, A. schubertii, A. siculum (syn. Nectaroscordum siculum), A. sphaerocephalon, A. strictum, A. stipitatum ‘Album’, A. ‘Ivory Queen’, A. ‘Lucy Ball’, A. ‘Mont Blanc’, and A. ‘Purple Sensation’. Scape length, inflorescence diameter, and flowering period were recorded. Isozyme marker polymorphism was assessed by starch gel electrophoresis. Eight polymorphic isozyme systems (AAT, GPI, PGM, ALAT, ACP, DIAP, ALDO, PGD) were selected from 16 analysed in the taxa. Besides the differences between the taxa, the isozymes revealed intraspecific polymorphism in 5 systems. A total of 37 markers were obtained and used for dendrogram construction. The most similar taxa were A. karataviense with A. ‘Ivory Queen’, and A. karataviense with A. christophii (similarity level 0.78). A high similarity of 11 taxa belonging to one group (A. aflatunense, A. christophii, A. giganteum, A. karataviense, A. nigrum, A. schubertii, A. ‘Ivory Queen’, A. ‘Lucy Ball’, A. stipitatum ‘Album’, A. ‘Mont Blanc’, A. ‘Purple Sensation’) suggested that this group could be identified with the subgenus Melanocrommyum.

Keywords: genetic similarity, intraspecific polymorphism, isozyme variation, ornamental alliums, phylogenetic analysis

Introduction morphism: apart from Allium and Amerallium also Caloscordum, Melanocrommyum and Rhizirideum The Allium encompasses 700 species and is (Samoylov et al. 1999). The subgenus Melano- distributed all over the Northern Hemisphere. crommyum was divided into 16 sections (Fritsch Apart from the familiar agricultural crops, many 1997). ornamental species belong to this genus, but its Three of the species of the subgenus is uncertain (Hong et al. 1996). Melanocrommyum included in this study were Three subgenera were separated on the basis of classified earlier by using GISH and RAPD morphological and anatomical characters: Allium, (Friesen et al. 1997). The RAPD technique was Amerallium and Nectaroscordum (Traub 1968, applied also by Hong et al. (1996) in the analysis 1972). A closely related species, A. siculum (syn. of ornamental alliums, including 5 species from Nectaroscordum siculum), having a characteristic the set of taxa evaluated in this study. In 1997, smell, is in some classifications excluded from the Dubouzet et al. analysed alliums from the subge- genus Allium because of the vascular bundle nus Rhizirideum (not present in our analysis) ac- pattern. Relatively recently, 5 subgenera were dis- cording to dot blot hybridization with randomly tinguished on the basis of chloroplast DNA poly- amplified DNA probes. Dubouzet and Shinoda

Received: February 20, 2007. Accepted: February 15, 2008. Correspondence: A. Krzymiñska, Department of Ornamental Plants, University of Life Sciences, D¹browskiego 159, 60–594 Poznañ, Poland; e-mail: [email protected] 214 A. Krzymiñska et al.

(1999) grouped Old and New World alliums ac- planted in rows in the first half of October. Each cording to ITS DNA sequence. Three species from taxon was planted in 3 rows, each containing this study were analysed in the above paper. 10 , so a total of 540 bulbs were planted. Isozymes were applied in Allium sativum L. by The bulbs originated from the Horticultural Farm Pooler and Simon (1993), Maass and Klaas of Mr Bogdan Królik (Poland), the Dutch com- (1995), and Ipek and Simon (2002). pany Toolen, and the Polish company Haubitz The aim of this study was to verify the similar- Polska. Cultivation conditions were equal for all ity of 13 species and 5 ornamental cultivars, and to plants. During the growing period, both morpho- classify them into particular groups based on mor- logical and phenological characters were ob- phological and isozyme variation. We report on served. The paper presents only selected the usefulness of isozymes for assessment of simi- characters, which are most important: scape larity of species and cultivars, in comparison to length, inflorescence diameter, and flowering morphological estimation. date. The first 2 characters were subjected to a one-way analysis of variance, while means were clustered by using Duncan’s test at a significance Materials and methods level of a = 0.05. GenStat software (GenStat 1993) was used for statistical analysis. Plant material Isozyme electrophoresis Our investigations covered 13 species and 5 cultivars of ornamental alliums: with purple flow- For the genetic analysis, 5 plants were chosen ran- ers, Allium aflatunense, A. christophii, A. giganteum, domly among the 30 plants of each taxon, which A. rosenbachianum, A. sphaerocephalon, A. ‘Lucy had been evaluated morphologically and Ball’, A ‘Purple Sensation’; with light pink flowers phenologically. The top sections (10-cm-long) of A. pyrenaicum, A. strictum, A. karataviense; with young (2–4-week-old) were collected in pink-purple flowers A. schubertii; with blue flowers 2003. Electrophoresis was run on 11% starch gels A. caeruleum; with yellow flowers A. moly; with (Gottlieb 1973). Fresh leaf materials for AAT white flowers A. nigrum, A. stipitatum ‘Album’, (aspartate aminotransferase), ALAT (alanine amino- A. ‘Ivory Queen’, A. ‘Mont Blanc’; and with yel- transferase), EST (esterase), GPI (glucosephosphate

Table 1. Taxonomic classification of evaluated Allium species and cultivars Species/cultivar Subgenus Section References A. aflatunense Melanocrommyum Megaloprason Linne von Berg et al. 1996 A. caeruleum Allium Caerulea Klass and Friesen 2002 A. christophii Melanocrommyum Kaloprason Linne von Berg et al. 1996 A. giganteum Melanocrommyum Megaloprason Dubouzet and Shinoda 1999 A. karataviense Melanocrommyum Miniprason Klass and Friesen 2002 A. moly Amerallium Molium Linne von Berg et al. 1996 A. nigrum Melanocrommyum Melanocrommyum Klass and Friesen 2002 A. pyrenaicum A. rosenbachianum Melanocrommyum Friesen et al. 1997 A. schubertii Melanocrommyum Kaloprason Hong et al. 1996 A. siculum Nectaroscordum Dubouzet and Shinoda 1996 A. sphaerocephalon Allium Allium Linne von Berg et al. 1996 A. strictum A. stipitatum 'Album' Melanocrommyum Megaloprason Linne von Berg et al. 1996 A. 'Ivory Queen' A. 'Lucy Ball' Melanocrommyum Friesen et al. 1997 A. 'Mont Blanc' A. 'Purple Sensation' Melanocrommyum Friesen et al. 1997

low-green flowers A. siculum. Table 1 presents their isomerase), LAP (leucine aminopeptidase), MDH classification according to published literature. (malate dehydrogenase), ME (malic enzyme), PGM Alliums were grown in the field in the (phosphoglucomutase), and PRX (peroxidase) anal- 2001/2002 and 2002/2003 seasons. Bulbs were yses were extracted in tris-maleate buffer. Isozymes in ornamental alliums 215

The potassium phosphate extraction buffer was used enzyme activity staining can be found at Wolko for ACP (acid phosphatase), ALDO (aldolase), and Œwiêcicki (1987). DIAP (diaphorase NADH), FUM (fumarase), IDH Polymorphism was interpreted as an absent or (isocitrate dehydrogenase), PGD (6-phospho- present band. The results were combined in a 0/1 gluconate dehydrogenase), and SKDH (shikimate data matrix and analysed by using GenStat soft- dehydrogenase). Fresh tissue was crushed, the crude ware (GenStat 1993). The coefficient of genetic extract was absorbed by filter paper wicks similarity was determined in the 18 ornamental (Whatman 31ET) and applied to the starch gel. Paper Allium taxa on the basis of the variability of 37 wicks were removed from the gel after 10 min of zones of enzymatic activity, defined as markers. electrophoresis. Four different buffer systems were The coefficients were calculated from the equation applied in order to optimise electrophoresis condi- (Nei and Li 1979): GAB =2NAB /(NA +NB), where tions for each analysed enzyme system. Samples N is the number of bands shared by objects A extracted by using tris-maleate buffer (Tris-amino- AB and B; N is the number of scored bands of object methane, maleic anhydride, polyvinylpyrrolidone A A; and N is the number of scored bands of b B (PVP-40), glycerol, TritonX-100, -merkapto- object B. The coefficients were used for hierarchi- ethanol) were placed on both A/B and T system. cal clustering by using the unweighted pair group The potassium phosphate extraction buffer method with arithmetic means (UPGMA). The re- (dipotassium hydrogen orthophosphate, hydro- lationships among species and cultivars were pre- chloric acid, contained the same components like sented in the form of a dendrogram. tris-maleate buffer) was used for samples sub- jected to electrophoresis on both H and C systems (Wolko 1995). System A/B was applied according Results to Selander et al. (1971), system T according to Weeden (1987), system C according to Clayton and Tretiak (1972), and system H according to The analysed Allium species and cultivars differed in Cardy et al. (1980). Electrophoresis was run at 5°C morphological characters (Table 2). The shortest in a refrigerator for 4–5 h. The electric current or scapes were found in A. ‘Ivory Queen’, potential at the starting point and after taking off A. karataviense, and A. moly (means 19.0–24.5 cm), the paper wicks was: 50 mA/300 V in system A/B; while the longest in A. rosenbachianum and A. ‘Lucy 40 mA/300 V in system C; 40 mA/250 V in system Ball’ (means 100.0–110.0 cm). The value of this H; and 25 mA/185 V in system T. The details of character in A. ‘Lucy Ball’ did not differ significantly

Table 2. Morphological and phenological features of Allium species and cultivars Species/cultivar Mean scape lengthMean inflorescence Flowering period (cm) diameter (cm) (weeks of year) A. aflatunense 81.9 gh 11.0 bc 19–22 A. caeruleum 46.4 cde 3.7 a 21–26 A. christophii 43.8 cd 24.5 e 19–26 A. giganteum 87.5 hi 9.2 bc 21–26 A. karataviense 21.5 ab 9.4 bc 19–23 A. moly 24.5 ab 5.8 a 22–25 A. nigrum 58.0 de 8.9 bc 21–23 A. pyrenaicum 38.0 bc 4.4 a 25–32 A. rosenbachianum 110.0 j 11.3 bc 19–22 A. schubertii 51.0 cde 34.0 f 22–27 A. siculum 76.5 fgh 15.5 d 21–25 A. sphaerocephalon 64.0 efg 3.8 a 24–30 A. strictum 60.0 def 5.7 a 26–32 A. stipitatum 'Album' 86.0 hi 11.5 c 21–25 A. 'Ivory Queen' 19.0 a 8.5 b 19–23 A. 'Lucy Ball' 100.0 ij 9.6 bc 19–22 A. 'Mont Blanc' 78.5 gh 9.9 bc 20–23 A. 'Purple Sensation' 77.0 fgh 10.5 bc 19–22 Mean values marked with the same letter do not differ at á =0.05, Duncan’s test (n = 30) 216 A.Krzymiñskaetal. from the value obtained in A. giganteum (mean 87.5 A. pyrenaicum, and in the 26th week in cm). One-way analysis of variance did not show sig- A. strictum. Long-blooming species included also nificant differences in scape length between A. christophii, whose flowering lasted from the A. giganteum, A. stipitatum ‘Album’, A. aflatunense, 19th tothe26th weekoftheyear. A. siculum, A. ‘Mont Blanc’, and A. ‘Purple Sensa- A total of 16 isozyme systems were analysed. tion’ (Table 2). No bands were observed for 5 systems (PRX, ME, The smallest inflorescence diameter was found in MDH, IDH, FUM). Three systems were not incor- A. caeruleum, A. sphaerocephalon, A. pyrenaicum, porated into the analysis due to a faint resolution A. strictum, and A. moly (means 3.7–5.8 cm). Mean (LAP, SKDH), or inconsistent enzyme activity or inflorescence diameter in A ‘Ivory Queen’, visualization with UV light (EST). Eight of the en- A. nigrum, A. karataviense, A. ‘Lucy Ball’, A. ‘Mont zyme systems provided consistently resolved, Blanc’, A. ‘Purple Sensation’, A.aflatunense, and polymorphic bands. These included GPT, ALDO: A. rosenbachianum ranged from 8.5 cm to 11.3 cm. 1 band in different species; PGM, ACP: 1 or 2 Values recorded for these alliums, except for bands; and AAT, PGD, DIAP, GPI: multiband A. ‘Ivory Queen’, did not differ significantly from in- patterns. A total of 37 zones of enzyme activity florescence diameter in A. stipitatum ‘Album’ (mean were analysed (namely Aat A, C–F; AcpA A–B; 11.5 cm). The largest diameter of inflorescences was AcpK A–D; Aldo A–D; Diap A–C; Gpi A–D; recorded in A. schubertii (mean 34.0 cm). Alat A–D; Pgd A–F; Pgm A, C–F). Bands on the Flowering date is a character frequently in- zymograms were scored as present (1) or absent cluded in descriptions of plant species. This char- (0), according to Ipek and Simon (2002) and Virk acter in the analysed taxa also varied. Allium et al. (2000) (Figure 1). Intraspecific mono- aflatunense, A. karataviense, A. rosenbachianum, morphic bands were visible in 3 enzyme systems. A. ‘Ivory Queen’, A ‘Lucy Ball’, and A. ‘Purple Five enzyme systems displayed heterozygous Sensation’ began flowering the earliest, in the 19th zones of isozyme activity. Thus the presence of week of the year (early May), and finished in the intraspecific Allium variation was detected, de- 22nd or 23rd week (early June). Flowers in spite vegetative propagation (in A. caeruleum, A. ‘Mont Blanc’ developed a week later. A. christophii, A. karataviense, A. moly, A. siculum, A. stipitatum ‘Album’, A. caeruleum, A. schubertii, A. siculum, A. strictum, A. ‘Ivory and A. giganteum started flowering in the 21st Queen’, A. ‘PurpleSensation’). week (late May) and the first 2 taxa finished to The similarity values ranged from 0.08 to 0.78, flower in the 25th week (mid-June), while the last with a value closer to 1.0 indicating greater similar- 2 in the 26th week (late June). Also A. nigrum be- ity (Table 3). The average similarity value was 0.4 gan flowering in the 21st week, but it lasted only with A.‘Ivory Queen’ showed 3 weeks. A short, 3-week flowering period was the highest affinity. The lowest similarity was de- also found in A. moly. In the other species flower- tected for A stipitatum ‘Album’ and A. pyrenaicum. ing lasted longer, 6 weeks or more. It began in the The dendrogram was constructed on the basis of the 22nd week in A. schubertii, in the 24th week in isozyme variation and hierarchical clustering by us- A. sphaerocephalon, in the 25th week in ing the UPGMA. Group I(A. pyrenaicum,

Figure 1. Isozyme patterns of the PGM enzyme system in Allium species and cultivars. Pgm A–D are analysed asmarkers 217

Table3.Coefficientsofgeneticsimilaritybetween Allium speciesandcultivars

A. caeruleum 0.42 A. christophii 0.32 0.10 A. giganteum 0.42 0.30 0.50 A. stipitatum 'Album' 0.45 0.17 0.52 0.43 A. karataviense 0.47 0.22 0.78 0.56 0.57 A. ‘Ivory Queen' 0.42 0.20 0.60 0.60 0.52 0.78 A. ‘LucyBall' 0.55 0.26 0.61 0.61 0.69 0.67 0.61 A. moly 0.22 0.11 0.53 0.42 0.55 0.47 0.42 0.36 A. ‘Mont Blanc' 0.36 0.26 0.61 0.52 0.54 0.76 0.70 0.69 0.45 A. nigrum 0.55 0.43 0.43 0.52 0.38 0.48 0.61 0.46 0.45 0.54 A. ‘PurpleSensation' 0.47 0.33 0.44 0.67 0.29 0.50 0.44 0.48 0.35 0.38 0.38 A. pyrenaicum 0.20 0.38 0.10 0.38 0.08 0.00 0.10 0.17 0.10 0.25 0.25 0.21 A. rosenbachianum 0.35 0.42 0.50 0.50 0.59 0.55 0.42 0.52 0.52 0.44 0.44 0.55 0.24 A. schubertii 0.56 0.21 0.53 0.74 0.36 0.59 0.63 0.64 0.44 0.55 0.64 0.59 0.20 0.35 A. siculum 0.55 0.35 0.43 0.26 0.62 0.48 0.52 0.46 0.45 0.38 0.46 0.38 0.17 0.59 0.27 A. sphaerocephalon 0.19 0.18 0.36 0.27 0.40 0.20 0.18 0.48 0.38 0.40 0.32 0.30 0.35 0.31 0.29 0.32 A. strictum 0.30 0.19 0.29 0.29 0.25 0.21 0.29 0.33 0.30 0.42 0.42 0.32 0.45 0.24 0.40 0.33 0.52 'Album' Ivory Queen' Lucy Ball' Mont Blanc' Purple Sensation' A. aflatunense A. caeruleum A. christophii A. giganteum A. stipitatum A. karataviense A.’ A. A. moly A. A. nigrum A. A. pyrenaicum A. rosenbachianum A. schuberti A. siculum A. sphaerocephalon

Figure2.Phylogeneticanalysisof Allium speciesandcultivars,basedon37isozymemarkers 218 A. Krzymiñska et al.

A. strictum, and A. sphaerocephalon) was separated A. sphaerocephalon showed morphological and from the other groups at the level of 0.28 within the phenological similarity. These alliums formed similarity tree (Figure 2). Groups II (A. moly, group I. For the first time, A. pyrenaicum and A. rosenbachianum, and A. siculum) and III deviated A. strictum were clustered under the group identi- at the level of 0.44. Group III formed 4 nodes: IIIa fied with the section Allium, subgenus Allium. (A. aflatunense and A. nigrum) and IIIb (A. ‘Purple This group displayed a lesser affinity with the sub- Sensation’, A. giganteum, and A. schubertii), with genus Melanocrommyum than with group II. This bifurcation at 0.5, as well as IIIc (A. ‘Lucy Ball’ and observation is consistent with the results of A. stipitatum ‘Album’) and IIId (A. ‘Mont Blanc’, Dubouzet and Shinoda (1999), who classified al- A. christophii, A. karataviense, and ‘Ivory Queen’), liums according to ITS DNA sequence. with bifurcation at 0.6. revealed a closer relation- ship with cultivars from group I, which was con- firmed by Hong et al. (1996). This species belongs Discussion to the subgenus Allium (Table 1). (representing the subgenus The isozyme data showed inter- and intraspecific Amerallium) and A. siculum (representing the sub- variability in alliums, higher than in other plant genus Nectaroscordum) formed group II. Also ac- species analysed earlier, like pea, field bean, lupin cording to Dubouzet and Shinoda (1999), (Œwiêcicki et al. 2000; Wolko et al. 2000). Six Pgd A. siculum was an outgroup. bands were observed in Allium cultivars, while le- Group III was identified with the subgenus gume species show only 2 Pgd loci. Intraspecific Melanocrommyum. Our data is contradictory to variation was observed for 5 isozyme systems Friesen et al. (1997) and Kamenetsky and Fritsch (AAT, GPI, PGM, ACP, PGD) in 50% of analysed (2002), because A. rosenbachianum (classified species. The presence of intraspecific isozyme under the subgenus Melanocrommyum) was an variation was unexpected due to the vegetative outgroup species in this study. Friesen et al. (1997) propagation of alliums. According to Hamrick verified Allium classification with GISH and et al. (1979), asexually reproducing plants exhibit RAPD. Kamenetsky and Fritsch (2002) used spe- less isozyme variation than sexually reproducing cific morphological characters and flowering pe- ones. Moreover, the most variable species are riod. A. rosenbachianum has rounded, relatively those capable of both sexual and asexual propaga- large inflorescences, similarly as all taxa included tion, which is not the case in alliums (Hamrick in group III, apart from A. nigrum. Moreover, out et al. 1979). Probably the reason of the variability of the 3 alliums included in group II, only in asexually propagating species is residual varia- A. rosenbachianum differed in flowering date. tion from ancestral sexually reproducing species. Subgroups IIIa, IIIb, IIIc, and IIId did not corre- In addition, somatic mutations can create spond to the sections proposed by Fritsch (1997). genotypic variation. Clonal offspring are expected Allium aflatunense and A. nigrum formed subgroup because of the lack of a mechanism of recombina- IIIa, while A. ‘Purple Sensation’ with A. giganteum tion in asexual organisms, but DNA can be and A. schubertii were clustered as subgroup IIIb. changed at various stages of plant development. The results did not confirm data on the origin of Mutations in undifferentiated meristematic tissue A. ‘Purple Sensation’ from A. aflatunense (Davies can be further incorporated into the germ line, so 1992). That hypothesis was also inconsistent with mutation in the parent is passed to the offspring the results of Friesen et al. (1997). Allium ‘Lucy (King and Schaal 1990). Allozymes usually dis- Ball’ and A. stipitatum ‘Album’ formed subgroup play insufficient polymorphisms among cultivars, IIIc, whereas A. ‘Ivory Queen’ displayed a close re- for example in Allium wakegi (Hong 1996). How- lationship to A. karataviense for the first time, and ever, in our study the variability in alliums, as- both were located within subgroup IIId. sayed with isozymes, was high enough for Allium ‘Mont Blanc’ was similar to A. christophii, similarity dendrogram construction. A. karataviense, and A. ‘Ivory Queen’. These taxa The isozyme variation was used for were also similar in terms of morphological and dendrogram construction based on hierarchical phenological characters. Our data did not confirm clustering with the UPGMA. Groups I, II, III were the suggestion that A. ‘Mont Blanc’ derives from separated within the dendrogram (Figure 2). Sub- A. stipitatum (Kamenetsky and Fritsch 2002). groups IIIa, IIIb, IIIc, and IIId within group III The purpose of this study was to confirm that were observed. A. pyrenaicum, A. strictum, and isozyme polymorphism, as compared with mor- Isozymes in ornamental alliums 219 phological and phenological data, enables making REFERENCES more reliable inferences on genetic similarity in alliums. Systematic inferences were made usually Cardy B, Stuber C, Goodman M, 1980. Techniques for starch gel electrophoresis of enzymes from maize from observations and comparisons of phenotypes (Zea mays L.). Dept. of Statistics Mimco, Series No. (Crawford 1983). However, several genes and the 137. Rayleigh: North Carolina State Univ. environment can influence the development of Clayton J, Tretiak D, 1972. Amine-citrate buffers for morphological characters. The assumption that pH control in starch gel electrophoresis. J Fish Res 2 plants lacking a feature are more similar to each Board Can 29: 1169–1172. other than to a third plant, produces an additional Crawford D, 1983. Phylogenetic and systematic infer- error. In the case of isozymes, alternative alleles ences from electrophoretic studies. In: Tanksley S, are always visible. The isozymes were historically Orton T, eds. Isozymes in plant genetics and breed- the first application of markers, but they still show ing. Part A. Amsterdam: Elsevier Science Pub- some advantages over other marker systems, e.g. lishers: 257–287. simplicity and speed of analysis. The reliability of Davies D, 1992. Alliums – the ornamental . Lon- don: B.T. Batsford. this method allows the verification of results by other researchers. The isozyme investigations of Dubouzet JG, Shinoda K, 1999. Relationships among Old and New World alliums according to ITS DNA large Allium collections showed the strength of sequence analysis. Theor Appl Genet 98: 422–433. this technique (Maass and Klaas 1995; Pooler and Dubouzet J, Shinoda K, Murata N, 1997. Phylogeny of Simon 1993). Allium L. subgenus Rhizirideum (G. Don ex Koch) Isozyme variation reflects only differences in Wendelbo according to dot blot hybridization with protein-coding genes. A question can be posed randomly amplified DNA probes. Theor Appl Genet whether protein polymorphism is a good predictor 95: 1223–1228. of overall levels of genetic diversity. In this study Friesen N, Fritsch R, Bachmann K, 1997. Hybrid origin of some ornamentals of Allium subgenus congruence was observed between the separated Melanocrommyum verified with GISH and RAPD. groups I, II, III and the taxa described in the litera- Theor Appl Genet 95: 1229–1238. ture. This seems to indicate the usefulness of Fritsch RM, 1997. Relations between west and central allozymes in genetic diversity assessment. Asian sections of Allium L. subg. Melanocrommyum (Webb et Berth.) Rouy. In: Ozturk MV, Secmen V, Gork G, eds. Plant life in southwest and central . Proceedings of 4th Plant Life of Southwest Conclusions Asia Symposium 21–28 May 1995, Izmir, Turkey: 128–140. A high intraspecific variation was detected in the GenStat, 1993. Genstat 5 Release 3 Reference Manual. analysed alliums. Five isozyme systems (AAT, GenStat 5 Committee, Oxford: Clarendon Press. GPI, PGM, ACP, PGD) exhibited intraspecific Gottlieb LD, 1973. Enzyme differentiation and phylog- variability in 50% of analysed taxa. Probably the eny in Clarkia franciscana, C. rubicanda and C. reasons of the variability are somatic mutations and/or amoena. Evolution 27: 205–214. residual variation from ancestral sexually reproducing Hamrick JL, Linhart YB, Mitton JB, 1979. Relation- ships between life history characteristics and species. A high similarity of 11 taxa, belonging to electrophoretically detectable genetic variation in group III (A. aflatunense, A. christophii, A. giganteum, plants. Am Rev Ecol Syst 10: 173–200. A. karataviense, A. nigrum, A. schubertii, A. ‘Ivory Hong C-J, Arisumi K, Etoh T, 1996. RAPD analysis of Queen’, A. ‘Lucy Ball’, A. stipitatum ‘Album’, ornamental alliums for phylogenetic relationship. A. ‘Mont Blanc’, A. ‘Purple Sensation’) suggests that Mem Fac Agr Kagoshima Univ 32: 51–58. this group could be identified with the subgenus Ipek M, Simon P, 2002. Evaluation of genetic diversity Melanocrommyum. A great distance between among allium clones using molecular markers: com- A. christophii and A. schubertii on the dendrogram parisonofAFLPs,RAPDandisozymes.Plant&Ani- indicates that the classification of both species to the mal Genome X Meeting, Posters: 5;http://www.ars.usda.gov/Research/docs.htm?doc section Kaloprason, subgenus Melanocrommyum is id=6357, 30th January 2007. uncertain. Observation of isozyme variability can be Kamenetsky R, Fritsch RM, 2002. Ornamental alliums. supplemented with observations of morphological In: Rabinowitch HD, Currah L, eds. Allium Crop characters in the classification of ornamental al- Science: Recent Advances. CABI Publishing, liums. Oxon: 459–491. King LM, Schaal BA, 1990. Genotypic variation within Acknowledgments. We would like to thank Mr asexual linkages of Taraxacum officinale. Proc Nat Bogdan Królik and the companies Toolen (the Neth- Acad Sci 87: 998–1002. erlands) and Haubitz Polska (Poland) for kindly pro- Klass M, Friesen N, 2002. Molecular markers in viding the Allium bulbs used in this study. Allium. In: Rabinowitch HD, Currah L, eds. Allium 220 A. Krzymiñska et al.

Crop Science – Recent Advances. CABI Pub- in the core collection of the Polish Pisum genebank. lishing, Oxon: 159–185. Genet Resour Crop Ev 47: 583–589. Linne von Berg G, Samoylov A, Klaas M, Hanelt P, Traub BH, 1968. The subgenera, sections and subsec- 1996. Chloroplast DNA restriction analysis and the tions of Allium L. Plant Life 24:147–163. infrageneric grouping of Allium (Alliaceae). Plant Traub BH, 1972. Genus Allium L. - subgenera, sections Syst Evol 200: 253–261. and subsections. Plant Life 28:132–137. Maass HI, Klaas M, 1995. Infraspecific differentiation Virk P, Zhu J, Newbury H, Bryan G, Jackson M, of (Allium sativum L.) by isozyme and RAPD Ford-Lloyd B, 2000. Effectiveness of different markers. Theor Appl Genet 91: 89–97. classes of molecular markers for classifying and re- Nei M, Li WH, 1979. Mathematical model for studying vealing variation in rice (Oryza sativa) germplasm. genetic variation in terms of restriction Euphytica 112: 275–284. endonucleases. Proc Nat Acad Sci 90: 10972–10976. Weeden NF, 1987. Observation of linkage between Pooler MR, Simon PW, 1993. Characterization and Wsp and isozyme locus Alat-p. Pisum Newslett 19: classification of isozyme and morphological varia- 80–81. tion in a diverse collection of garlic clones. Wolko B, 1995. Markery molekularne w badaniach Euphytica 68: 121–130. genetycznych rodzaju Lupinus. [Molecular markers Samoylov A, Friesen N, Poolner S, Hanelt P, 1999. Use in genetic studies of the genus Lupinus]. Hodowla of chloroplast DNA polymorphism for the study of Roœlin i Nasiennictwo 39: 3–64. Allium subgenus Amerallium and subgenus Wolko B, Œwiêcicki WK, 1987. The application of Bromatorrhiza (Alliaceae) II. Feddes Repert: electrophoretic methods of isozymes separation for 103–109. genetical characterization of pea cultivars. Genet Selander RK, Smith MH, Yang SY, Johnson WE, Gen- Pol 26: 89–99. try JB,1971. Biochemical polymorphism and sys- Wolko B, Œwiêcicki W, Kruszka K, Irzykowska L, tematics in the genus Peromyscus. I. Variation in 2000. Isozyme and RAPD markers for the identifi- the old-field mouse (Peromyscus polionotus). Univ cation of pea, field bean and lupin cultivars. J Appl Tech Publ 7103: 49–90. Genet 41: 151–165. Œwiêcicki W, Wolko B, Apisitwanich S, Krajewski P, 2000. An analysis of isozymic loci polymorphism