Hybridization at intraspecific, interspecific, and intergeneric levels is a well-known breeding strategy to create variations with desirable traits for ornamental flowers. A total of 161 crossing combinations were made on three taxonomic levels, including 12 intraspecific crossing combinations within L. chalcedonica and L. ×haageana, 102 interspecific crossing combinations within Lychnis spp., and 47 intergeneric crossing combinations between Lychnis spp. and Silene spp. Intraspecific crosses showed high cross-compatibility, which yielded mature seeds and progeny plants. Most of the interspecific crossings in genus Lychnis produced limited seed set and germination percentages; however, L. cognate, L. ×arkwrightii, and L. ×haageana showed high cross-compatibility and might be more closely related than other species in Lychnis. As a result of cross-incompatibility, crossing combinations between Lychnis spp. and Silene spp. produced few flowers that set seeds. Significant differences occurred in seed set between crossing combinations and their reciprocal crosses for interspecific and intergeneric crosses. For these hybrids with immature seeds, embryo rescue techniques with immature seed culture would be the only way to produce seedlings.
The genera Lychnis L. and Silene L. belong to the family Caryophyllaceae. Lychnis species, commonly known as campion or catchfly, are native to the temperate regions of the Northern Hemisphere, from East Asia to Central Asia, Europe, and North and East Africa (Oxelman et al., 2000; Popp et al., 2008). Lychnis is characterized by perennial herbs, consisting of ≈30 species with some species being extensively cultivated as landscape ornamentals. Silene is the largest genus in Caryophyllaceae, consisting of biennial or perennial herbs with species numbers varying from 500 to 700 (Melzheimer, 1988; Oxelman et al., 2000; Walters, 1989). Silene species occur worldwide, although the greatest diversity occurs in the Mediterranean region (Oxelman et al., 2000). The taxonomic boundary between Lychnis and Silene L. is often disputed (Desfeux and Lejeune, 1996; Kruckeberg, 1962). Both genera have been merged in cladistic theory (Desfeux and Lejeune, 1996; Negrutiu et al., 2001), yet Lychnis as an independent genus is also accepted by researchers beyond taxonomic fields. Distinction between the traditional genera Lychnis and Silene in morphology is that Lychnis usually has five styles and an entire capsule, whereas Silene has three styles and a split capsule (Desfeux and Lejeune, 1996).
Lychnis has ornamental merit because most species are perennial, have good drought tolerance, and exhibit high diversity in morphology and ecophysiology. The majority of Lychnis and Silene species have identical somatic chromosome numbers (2n = 2x = 24) (Negrutiu et al., 2001), which suggests the possibility of hybridization for exploring desirable and exotic hybrids for commercial use. Hybridization has been used to test the phylogeny relationship of Lychnis and Silene (Kruckeberg, 1962; Wilson et al., 1995). A number of cultivars are commercially available and were obtained from intraspecific breeding (Godo et al., 2009; Mayol and Rossello, 2006). An exceptional example is Lychnis ×arkwrightii Heydt, which is a putative hybrid between Lychnis chalcedonica L. and Lychnis ×haageana Lemoine. The majority of Lychnis species are hermaphrodites and are cross-pollinated by insects, which is aided by being protandrous, although self-pollination does occur. Therefore, crossing combinations were made to test the crossability among Lychnis and Silene species and cultivars.
Materials and Methods
Plant materials and growth conditions.
Seven Lychnis species, totaling 20 accessions including commercial cultivars, hybrids, and species, and five Silene species were used for hybridizations (Table 1). A total of 161 crossing combinations were made on three taxonomic levels including: 1) 12 intraspecific crossing combinations within L. chalcedonica and L. ×haageana; 2) 102 interspecific crossing combinations within Lychnis spp.; and 3) 47 intergeneric crossing combinations between Lychnis spp. and Silene spp. Five crosses including reciprocal crosses were attempted for each combination without emasculation, yet the actual crossing numbers were restricted to flower availability. Overall, 49 crosses were made for intraspecific crossing, 623 crosses were made for interspecific crossing, and 290 crosses were made for intergeneric crossing. A total of 962 crosses were made for all combinations. To facilitate the large number of hybridizations, flowers were not emasculated; however, all progeny were evaluated based on flower and plant morphological traits to determine if progeny was the result of selfing. Progeny was removed from the study if they were a result of selfing. Fertilized capsules were collected as the color changed from green to brown. All hybridizations were conducted during May to Oct. 2010 in the Department of Horticulture and Landscape Architecture Research Greenhouses, Oklahoma State University, Stillwater, OK, under natural photoperiods. Temperature was set at 18/21 °C day/night with a photosynthetic photon flux density range of 600 to 1200 μmol·m−2·s−1 at 1200 hr. Seeds from individual capsules were sown in Metro-mix 702 (Sun Gro Horticulture, Bellevue, WA) media placed in 15-cm round pots (Itml, Middlefield, OH) starting July 2010 in the greenhouse.
Lychnis and Silene species and cultivars used in hybridization and their seed sources.
Beginning in July 2010, the number of crosses, number of successful crosses, F1 seed number, and germination rates were recorded. Phenotypic traits including leaf color, flower color, and size were recorded starting in May 2011. Hybrid flower morphological data of corolla diameter, petal width, and petal length were recorded on five flowers, each of them from an individual plant. Flower and leaf color were recorded using The Royal Horticultural Society Color Charts (1966). All hybrid plants were cultivated in the greenhouse when the color data were taken. The traits were analyzed using analysis of variance (PROC GLM, Version 9.3; SAS®, Cary, NC), and means were compared using a Duncan test (P ≤ 0.05, n = 5).
All 12 of the intraspecific combinations within Lychnis species set seed (Tables 2 and 3). Among six accessions within L. chalcedonica and 48 pollinated flowers, on average 83.3% set seed and had an average seed number of 52 per capsule. A total of 2094 seeds was collected from intraspecific crosses within L. chalcedonica, and average germination rate was 64.7% (Table 2). Only one intraspecific cross between L. ×haageana ‘Molten Lava’ and L. ×haageana mixed hybrids was made, of which 24 seeds were collected from the single capsule and 91.7% of them germinated normally (Tables 2 and 3). These results confirmed that interbreeding within species could successfully produce fertile seed and progeny. For L. chalcedonica, L. chalcedonica var. alba, and L. chalcedonica ‘White’, differences in percent of flowers that set seeds and seed germination rates within reciprocal crosses occurred (Tables 2 and 3). In particular, for a cross combination between L. chalcedonica ‘White’ and L. chalcedonica, seed germination rate was 84.8% when L. chalcedonica ‘White’ as a maternal parent, much higher than that of 22.0% as a paternal parent. Red flower was dominant in all intraspecific F1 hybrid plants among L. chalcedonica (red), L. chalcedonica var. alba, L. chalcedonica ‘White’, and L. chalcedonica ‘Rauhreif’ (white), or L. chalcedonica ‘Carnea’ (dusty pink) flowers (data not shown).
Flower pollination, seed set, and germination for intraspecific crosses as seed parent within Lychnis.
Flower pollination, seed set, and germination for intraspecific crosses as pollen parent within Lychnis.
Compared with the intraspecific crosses, most interspecific crosses in the genus Lychnis produced fewer flowers that set seed and had lower seed germination percentages (Tables 4 and 5). However, the percent of flowers that set seed and germinated was relatively high for the crossing pairs between L. cognata and L. ×arkwrightii ‘Vesuvius’, and L. ×haageana (L. ×haageana mixed hybrids and ‘Molten Lava’, ‘Lumina Orange’, and ‘Lumina Broze leaf Red’), and also between L. ×arkwrightii ‘Vesuvius’ and L. ×haageana mixed hybrids. In the crossing pairs where L. cognata was the maternal parent, 57.1% of crosses with L. ×arkwrightii ‘Vesuvius’ and 80.0% of crosses with L. ×haageana mixed hybrids produced seed and seed germination rates were 77.1% and 68.5%, respectively (Table 6). In crosses with L. ×arkwrightii ‘Vesuvius’ as the maternal parent and L. cognate as the pollen parent, seed set and seed germination rates were 93.8% and 69.2%, respectively. Seed set was lower for the crossing pairs with L. ×haageana mixed hybrids where L. cognate was the pollen parent; however, seed germination rates were similar (Table 6). In the reciprocal crosses between L. ×haageana ‘Molten Lava’ and L. cognate, and the crosses between L. ×haageana ‘Lumina Orange’ or ‘Lumina Bronze Leaf Red’ with L. cognate, 100.0% of crosses produced seeds, and seed germination rates ranged from 53.2% to 87.3% (Table 6). For L. ×arkwrightii ‘Vesuvius’ × L. ×haageana mixed hybrids, 40.0% of pollinated flowers set seed and 60.7% of the seeds germinated. Meanwhile, the reciprocal crosses between the two accessions yielded 87.5% seed setting flowers and 68.2% seed germination (Table 6). The results demonstrated that no interbreeding barrier exists within the three species including L. cognate, L. ×arkwrightii, and L. ×haageana out of the seven species we tested in genus Lychnis. They are likely more closely related to each other than the other species.
Flower pollination, seed set, and germination for interspecific crosses excluding crosses in Table 6 as the seed parent within Lychnis.
Flower pollination, seed set, and germination for interspecific crosses excluding crosses in Table 6 as the pollen parent within Lychnis.
Flower pollination, seed set, and germination for interspecific crosses between Lychnis cognata and L. ×arkwrightii, or L. ×haageana.
For the other 403 interspecific crossing pairs in Lychnis, except for L. miqueliana × L. ×arkwrightii ‘Orange Genome’, all of the interspecific crossing pairs produced seeds (Tables 4 and 5). Considering the average data for each species, L. flos-cuculi and L. wilfordii produced relatively few seed setting flowers, seeds per capsule, and number of germinated seeds. The results indicated that some interspecific cross-compatibility barriers prevented hybridization and obtaining normal seeds from crossing combinations including L. chalcedonica × L. cognata, L. chalcedonica × L. miqueliana, L. chalcedonica × L. ×arkwrightii, L. chalcedonica × L. ×haageana, L. wilfordii × L. chalcedonica, L. cognate × L. flos-cuculi, L. cognate × L. miqueliana, L. cognate × L. wilfordii, L. miqueliana × L. wilfordii, L. miqueliana × L. ×arkwrightii, L. flos-cuculi × L. miqueliana, L. miqueliana × L. ×haageana, L. flos-cuculi × L. ×arkwrightii, L. wilfordii × L. ×arkwrightii, L. wilfordii × L. ×haageana, L. wilfordii × L. flos-cuculi, and L. flos-cuculi × L. ×haageana. Nevertheless, because most of the crossing combinations produced immature seeds, embryo rescue techniques with immature seed culture might overcome the barriers and produce hybrid seedlings.
For the interspecific crosses among Lychnis species, there were significant differences for seed set between crossing combinations and their reciprocal crosses (Tables 4, 5, and 6). Therefore, crossing two interspecific parents in both directions would be a practical method to partially overcome cross-incompatibility between two species in Lychnis. F1 hybrid plants from 11 cross-compatible combinations among L. cognate, L. ×arkwrightii, L. ×haageana, and from L. ×haageana ‘Molten Lava’ × L. miqueliana had intermediate ornamental traits like flower color, stem color, stem hardness, and leaf color between both parents (Tables 7 and 8). These phenotypes showed that promising hybrid plants could be obtained from interspecific crosses, although their hybrid identities need to be evaluated further by molecular marker methods or by observing segregation patterns in F2 generation. For L. ×arkwrightii ‘Vesuvius’ × L. chalcedonica var. alba, L. chalcedonica var. alba × L. ×haageana mixed hybrids, and L. ×haageana mixed hybrids × L. chalcedonica var. alba, progeny flower color was more similar to the maternal parent.
Flower color in parents and their hybrids for interspecific crosses in Lychnis.
Morphological assessments for six selected interspecific hybrids in Lychnis.
Because of cross-incompatibility, crossing combinations between Lychnis spp. and Silene spp. had the lowest number of seed setting flowers and seeds. Of 290 crosses made, only 39 set seed (Tables 9 and 10). Partial compatible crossing combinations were made from L. chalcedonica × S. armeria, L. cognate × S. plankii and its reciprocal cross, L. cognate × S. pendula, S. plankii × L. miqueliana, L. ×arkwrightii ‘Vesuvius’ × S. armeria, L. ×arkwrightii ‘Vesuvius’ × S. pendula, S. plankii × L. ×arkwrightii ‘Vesuvius’, L. ×haageana mixed hybrids × S. armeria, L. × haageana mixed hybrids × S. pendula, L. × haageana ‘Molten Lava’ × S. pendula, L. wilfordii × S. armeria, and L. wilfordii × S. pendula. Among intergeneric crossing combinations, seed set rates varied. No seeds germinated; thus, embryo rescue techniques with immature seed culture would likely have to be used to produce seedlings. Again, seed set differences in reciprocal crosses helped to partially overcome interspecific cross-incompatibility between two genera.
Flower pollination, seed set, and germination for intergeneric crosses as seed parent between Lychnis and Silene.
Flower pollination, seed set, and germination for intergeneric crosses as pollen parent between Lychnis and Silene.
According to our observation, L. chalcedonica and L. flos-cuculi have a very high selfing rate. Lychnis wilfordii yielded a very low selfing rate in the greenhouse despite producing a lot of pollen. No hybrid plants were obtained from L. wilfordii, yet the profuse flowering makes it a potential parent for ornamental breeding through embryo rescue.
The common name of the family Caryophyllaceae is carnation family or pink family; however, pink color is not a prevailing color in Lychnis species used in the hybridizations. Two species, L. miqueliana and L. cognata, are pink with relatively large flowers making them good potential parents, but both had weak stems when cultivated in the greenhouse. Lychnis miqueliana did not readily produce hybrids with other species. The interspecific hybridizations between L. cognata and other Lychnis species were successful. Six hybrids generated from L. cognata were pink (Table 7). The stem color of the hybrids between L. ×arkwrightii ‘Vesuvius’ and L. cognata was partially inherited from ‘Vesuvius’ showing reddish internodes. The stems were stronger than L. cognata. Lychnis cognata proved to be a good parent for crossing for its promising interspecific cross-compatibility and pink flower color. In this study, at least 11 interspecific hybrids with rich variations in flower color, stem type, stem color, and leaf color were generated without the need for embryo rescue. Six promising hybrids were selected according to their attractive ornamental phenotypes and morphology (Table 8). To expand genetic variations for ornamental flower breeding programs using Lychnis and Silene, immature seed culture protocols, phenotype selection, fertility status, and propagation protocols for hybrids need to be investigated further.
Hybridization at intraspecific, interspecific, and intergeneric levels is commonly practiced in breeding programs to combine desirable traits. Our results demonstrated that mature seeds and normal plants can be produced from intraspecific hybridization within L. chalcedonica and L. ×haageana species, confirming results by Nakano et al. (2013). In their study, Nakano et al. (2013) made cross-pollinations among eight species with the same chromosome number (2n = 2x = 24) in Lychnis, including L. chalcedonica, L. coronata, L. fulgens, L. gracillima, L. kiusiana, L. miqueliana, L. sieboldii, and L. wilfordii; based on the average value of each species, L. coronata yielded both the lowest seed set (8.3%) and seed number with four per capsule. Seed set for the other seven species ranged from 50.6% to 81.5%, and seed number per capsule ranged from seven to 55. However, only immature seeds were obtained from these interspecific crossing combinations 4 weeks after pollination. By immature seed culture on half-strength Murashige and Skoog medium without plant growth regulators, of a total of 26 crosses that produced seeds, seeds from 11 crosses germinated in 3 months, and only one of them produced seedlings. Compared with the previously mentioned report, we used different species and many cultivars in some species to test crossing compatibility for interspecific crosses in Lychnis. Also, the number of crosses including reciprocal crosses in our study was higher. Based on our study, high seed set and germination percentages were observed from the interspecific crossing combination among L. cognate, L. ×arkwrightii, and L. ×haageana with no cross-incompatible barriers existing between them. For most of the other interspecific crosses, our results were similar to Nakano et al.’s report (Nakano et al., 2013). For those crosses that only have immature seeds, embryo rescue methods need to be developed to produce hybrid plants. Also, the fertility and propagation procedure of interspecific hybrids should be investigated further.
Although all plants have the same chromosome number, cross-compatibility and hybrid seed germination between Lychnis and Silene were significantly lower than those between species in the genus Lychnis. The production of F1 hybrids showed some indication that Lychnis and Silene are closely related. This breeding evidence seems to support the conclusion that Lychnis and Silene are two genera instead of one genus classification (Rabeler, 1992). However, considering the large number of species in Silene and their broad distribution, the generic delimitation of Silene still needs to be studied in detail.
Because a significant number of crosses in a blooming season was performed with limited personnel in this study, hand hybridization without emasculation was conducted to save time and labor. Crosses are sometimes made without emasculation such as Yang et al. (2010) who conducted intergeneric hybridization without emasculation between Opisthopappus taihangensis (Ling) C. Shih. and Chrysanthemum lavandulifolium (Fisch. ex Trautv.). From our observation in the greenhouse, Lychnis is protandry, which facilitates non-emasculation if large hybrid populations are wanted with limited personnel. Genotype fingerprinting tools could be used to verify hybridity (Nakano et al., 2013). Pollen storage and embryo rescue techniques would facilitate a greater number of intraspecific, interspecific, and intergeneric crosses and hybrids.
Desfeux, C. & Lejeune, B. 1996 Systematics of euromediterranean Silene (Caryophyllaceae): Evidence from a phylogenetic analysis using ITS sequences Cr. Acad. Sci. Iii-Vie 319 351 358
Godo, T., Miyazaki, J., Kuwayama, S., Ogita, S., Kato, Y., Nakano, M. & Nakata, M. 2009 Interspecific hybridization between triploid Senno (Lychnis senno Siebold et Zucc., Caryophyllaceae) and allied taxa of the genus Lychnis Plant Biotechnol. 26 301 305
Melzheimer, V. 1988 Silene. In: Rechinger, K.H. (ed.). Flora iranica Lfg. Akademische Druck- & Verlagsanstalt, Graz. 163:341–508
Nakano, M., Kuwayama, S., Oka, E., Asano, M., Han, D. & Godo, T. 2013 Cross-compatibility in interspecific hybridization of Lychnis (Caryophyllaceae) and characterization of interspecific hybrids between L. fulgens and L. sieboldii J. Jpn. Soc. Hort. Sci. 82 57 62
Negrutiu, I., Vyskot, B., Barbacar, N., Georgiev, S. & Moneger, F. 2001 Dioecious plants. A key to the early events of sex chromosome evolution Plant Physiol. 127 1418 1424
Oxelman, B., Lidén, M., Rabeler, R.K. & Popp, M. 2000 A revised classification of the tribe Sileneae (Caryophyllaceae) Nord. J. Bot. 20 743 748
Popp, M., Gizaw, A., Nemomissa, S., Suda, J. & Brochmann, C. 2008 Colonization and diversification in the African ‘Sky Islands’ by Eurasian Lychnis L. (Caryophyllaceae) J. Biogeogr. 35 1016 1029
Royal Horticultural Society. 1966. Royal Horticultural Society colour chart. Royal Hort. Soc., London, and Flower Council of Holland, Leiden.
Walters, S.M. 1989 Silene; Lychnis, p. 179–182. In: Walters, S.M., J.C.M. Alexander, A. Brady, C.D. Brickell, J. Cullen, P.S. Green, V.H. Heywood, V.A. Matthews, N.K.B. Robson, P.E. Yeo, and S.G. Knees (eds.). The European garden flora, 3. Cambridge University Press, Cambridge, UK
Wilson, G.B., Wright, J., Lusby, P., Whittington, W.J. & Humphries, R.N. 1995 Lychnis viscaria L (Viscaria vulgaris Bernh) J. Ecol. 83 1039 1051
Yang, D.Y., Hu, X., Liu, Z.H. & Zhao, H.E. 2010 Intergeneric hybridizations between Opisthopappus taihangensis and Chrysanthemum lavandulifolium Sci. Hort. 125 718 723