Photomicrographs of condensed stained chromosomes of Theaceae. (A) Gordonia lasianthus 2006-220, 2n 2x = 36. (B) Schima superba 2018-009, 2n = 2x = 36. (C) Stewartia pseudocamellia 2018-111, 2n = 2x = 34. (D) Camellia japonica ‘Dr. JC Raulston’ 2017-060, 2n = 2x = 30. (E) Camellia sinensis ‘Red Leaf’ 2017-111, 2n = 2x = 30. (F) Camellia azalea 2018-063, 2n = 2x = 30. (G) Camellia ×vernalis ‘Egao Corkscrew’ 2017-062, 2n = 4x = 60. (H) Polyspora chrysandra 2015-114, 2n = 2x = 30. (I) Pyrenaria spectabilis 2018-008 2n = 2x = 30.
Click on author name to view affiliation information
Camellia L., the most speciose member of the diverse tea family Theaceae, has a long and complex horticultural history. Extensive cultivation and hybridization have produced thousands of varieties of Camellia, including commercially important crops such as cultivated tea, oilseed, and iconic flowering shrubs. Cytogenetics of Camellia and related genera is complicated; chromosome number and ploidy can vary widely between species, and interspecific and interploid hybridization occurs. However, specific information regarding cytogenetics of many species, cultivars, and modern hybrids is lacking. The objectives of this study were to compile a consolidated literature review of the cytogenetics of Camellia and related genera and to determine chromosome numbers, ploidy, and genome sizes of specific accessions of selected species, cultivars, and interspecific and interploid hybrids. A review of the existing literature regarding Theaceae cytogenetics is presented as a consolidated reference comprising 362 taxa. Genome sizes were determined with flow cytometry using propidium iodide as a fluorochrome and Pisum sativum ‘Ctirad' and Magnolia virginiana ‘Jim Wilson’ as internal standards. Chromosome numbers of selected taxa were determined using traditional cytology and were used to calibrate genome sizes with ploidy level. Our results confirmed a base chromosome number of x = 15 for Theeae including Camellia, x = 17 for Stewartiae, and x = 18 for Gordoniae. Surveyed camellias ranged from 2n = 2x = 30 to 2n = 8x = 120, including diploids, triploids, tetraploids, pentaploids, hexaploids, and octoploids. Previously uncharacterized taxa such as Camellia azalea, C. amplexicaulis, C. chrysanthoides, C. cordifolia, C. cucphuongensis, C. flava, C. nanyongensis, and C. trichoclada were found to be diploid. Ploidy was also newly determined for Schima argentea, S. khasiana, S. remotiserrata, and S. sinensis (all diploids). Both diploid and triploid Stewartia ovata were found, and a ploidy series was discovered for Polyspora that ranged from diploid to octoploid. Ploidy determinations were used to confirm or challenge the validity of putative interploid hybrids. Monoploid genome sizes varied among subfamily and genera, with 1Cx values ranging from 0.80 pg for Franklinia to a mean of 3.13 pg for Camellia, demonstrating differential rates of genome expansion independent of ploidy. Within Camellia, monoploid genome sizes varied among subgenera, sections, and some species (range, 2.70–3.55 pg). This study provides a consolidated and expanded knowledgebase of ploidy, genome sizes, hybridity, and reproductive pathways for specific accessions of Camellia and related genera that will enhance opportunities and strategies for future breeding and improvement within Theaceae.
Theaceae (Mirb. ex Ker Gawl.), the tea family, is a small family of trees and shrubs with a disjunct eastern Asian–eastern North American and northern South American distribution (Stevens, 2001 onwards). Camellia L. is the largest and most commercially significant genus of Theaceae, with species found throughout southeastern and eastern Asia (Chang and Bartholomew, 1984; Luna Vega and Contreras-Medina, 2000). Approximately 90% of all Camellia species, including those of greatest commercial importance, are native to China and Japan (Bartholomew, 1986). The ornamental varieties are prized for their glossy evergreen foliage and abundant showy flowers that can bloom from autumn to early spring, when many other plants in the landscape are dormant. More than 1000 years before their western introduction, ornamental camellias were grown for garden use in China (Xin et al., 2015). Although tea (C. sinensis) arrived in Europe during approximately the middle of the 17th century, the first living Camellia plant was not reported until nearly one century later, in Lord Petre Thorndon’s hothouses in England. Since then, ornamental camellias have become widely cultivated throughout Europe, North America, Australia, and New Zealand (Ackerman, 2007; Darfler, 2014; Trehane, 2007). Their popularity and phenotypic variability have led to tens of thousands of cultivars and hybrids (International Camellia Society, 2015). However, there are many polyploid camellias, and many species and complex hybrids have not been analyzed for ploidy or genome size. Improved knowledge of chromosome numbers and ploidy levels of key species and cultivars would be a valuable resource for further breeding and improvement of Camellia. Analyses of other closely related genera would provide a broader understanding of ploidy within Theaceae and help contextualize evolutionary relationships in this family.
The genus Camellia has undergone several taxonomic revisions (Prince, 2007). Sealy (1958) published a revision of the genus Camellia that included 12 sections and 82 species with an additional group of 24 doubtful species. Chang and Bartholomew (1984) and Chang (1998) completed several taxonomic revisions of Camellia and reorganized the 238 species native to China in 18 sections and 4 subgenera. More recently, Ming (2000) published his monograph of the genus Camellia; in that work, he reduced the number of subgenera to 2, the sections to 14, and the species to 119, which was less than half of Chang’s final tally of ≈280 species. The systems of both Ming (2000) and Chang (1998) are widely used by botanists today (Gao et al., 2005). In the work by Ming and Bartholomew (2007), the number of species is ≈120, and 97 of these are native to China.
Several changes to the accepted taxonomic status of Camellia have occurred since its original classification by Linnaeus. At different times, Camellia has been placed with the Guttiferales, Theales, and even within Ternstroemiaceae (Luna and Ochoterena, 2004). Theaceae is now considered a distinct family within Ericales (Stevens, 2001 onwards). Luna and Ochoterena (2004) found that Theaceae is closely related to Ternstroemiaceae; both belong to the same clade. Within Theaceae, there are three tribes: Theeae, Gordoniae, and Stewartiae. Theeae is the most diverse of these tribes and contains Camellia, Polyspora, Pyrenaria, Apterosperma, and Laplacea. A recent molecular phylogenetic analysis of Theeae indicated that Camellia and Pyrenaria form a paraphyletic group; these two occur and hybridize naturally with each other (Zhang et al., 2014). Gordoniae and Stewartiae belong to the sister clade of Theeae. Gordoniae is composed of the North American Franklinia and Gordonia, as well as the Asian Schima. Intergeneric hybrids between these species as well as between Camellia and Franklinia have been reported (Ackerman and Williams, 1982; Orton, 1977; Ranney et al., 2003; Ranney and Fantz, 2006). Stewartiae includes the disjunct North American and Asian Stewartia (including evergreen species sometimes classified as Hartia) (Prince and Parks, 2001).
There is some variation in the base chromosome number among members of Theaceae. Many genera including Camellia, Polyspora, and Pyrenaria exhibit base chromosome numbers consistently reported as 1n = 1x = 15 (Kondo, 1977; Yang et al., 2000, 2004). Other genera, however, have inconsistent reports, such as Stewartia with 1n = 1x = 15 or 17 and Franklinia, Gordonia, and Schima with 1n = 1x = 15 or 18 (Bostick, 1965; Horiuchi and Oginuma, 2001; Oginuma et al., 1994; Santamour, 1963). For Camellia, there is also considerable variability in ploidy, both among and within species. For example, Camellia japonica is most commonly found to be diploid (Ackerman, 1971; Kondo, 1977), although triploids, tetraploids, pentaploids, and aneuploids have been reported (Fukushima et al., 1966; Kondo, 1977). Although Camellia sasanqua is often reported to be hexaploid (Ackerman, 1971; Kondo, 1977), pentaploids, heptaploids, octoploids, decaploids, and aneuploids have been noted (Ito et al., 1957; Kondo, 1977). Ploidy series are seen in other Camellia species as well, including (but not limited to) C. hiemalis, C. oleifera, C. reticulata, and C. sinensis (Ackerman, 1971; Bezbaruah, 1971; Huang et al., 2013; Kondo, 1977). The variation and confusion regarding ploidy levels are further complicated by and may be partly the result of interspecific and interploid hybridization. For example, advanced hybrids of C. ×vernalis (C. sasanqua ×japonica) can be triploid, tetraploid, pentaploid, or hexaploid (Tateishi et al., 2007). Additionally, camellias can produce both unreduced gametes (Wendel, 1984) and, in some instances, aneuploid gametes (Kondo, 1977), resulting in additional possible variations within ploidy.
Ploidy and genome size can influence reproductive compatibility, fertility, and heritability of traits. Relative ploidy levels among related taxa can reflect and help elucidate biodiversity, genomic evolution, and taxonomic relationships (Laport and Ng, 2017; Ranney et al., 2018; Soltis et al., 2015). For example, seed development from interploid crosses can be limited by the failure of endosperm formation, leading to the production of nonviable seeds (Ramsey and Schemske, 1998). Anisoploid plants, whose chromosome numbers are in odd multiples of their basic number (e.g., triploid, pentaploid, etc.), can be sterile or have greatly reduced fertility, thereby limiting their potential as breeding lines. With increasing ploidy, allelic segregation becomes more complicated, thus leading to complex patterns of heritability, especially in autopolyploids (Zielinski and Scheid, 2012). Information regarding ploidy and genome size can also be used to confirm interploid hybridity, and genome size data can be used to estimate ploidy among related taxa when calibrated with known cytological standards.
The objectives of this study were to 1) conduct an extensive literature review and compile a consolidated reference regarding the cytogenetics of Camellia and related genera, and 2) to augment prior research with original data regarding ploidy and genome sizes of specific accessions of selected species, cultivars, and interspecific and interploid hybrids.
Tissue samples of species, hybrids, and cultivars of Camellia and closely related genera were collected from nurseries, private collections, and botanic gardens. Several species not previously reported were surveyed, and putative interploidy and interspecific hybrids were verified or challenged. Taxa with variable reported genome sizes were analyzed to determine the ploidy of specific clones. Cultivars with previously determined chromosome numbers were included to calibrate genome size with ploidy.
Flow cytometry was used to determine genome sizes following the methods of Huang et al. (2013). Approximately 40 to 50 mg of leaves were used for each sample preparation. A modified woody plant buffer (WPB) isolation buffer composed of 0.2 mm Tris-HCl, 4 mm MgCl2-6 H2O, 2.0 mm EDTA Na2-H2O, 86.0 mm NaCl, 2.0 mm dithiothreitol, 1% (w/v) PVP-10, and 1% (v/v) Triton X-100 1 mL with a pH of 7.5 was prepared. Ice-cold nuclei suspensions were prepared by chopping tissue in the WPB with a razor blade. The WPB buffer was used to reduce the effects of phenolic compounds, preserve chromatin integrity in the DNA, and help produce low cv values (Huang et al., 2013). The suspensions were filtered through a 50-μm nylon filter. The nuclei were subsequently treated with 50 μg·mL−1 RNase and stained with propidium iodide (PI) (Huang et al., 2013). Pisum sativum ‘Ctirad’ (2C = 8.75 pg) and Magnolia virginiana ‘Jim Wilson’ (2C = 3.92 pg) were used as internal standards. Samples were analyzed using a Partec PA II flow cytometer (Partec, Görlitz, Germany) to determine genome size. Holoploid, somatic, sporophytic, unreduced 2C genome size was calculated as the DNA content of the standard (pg) × (mean fluorescence value of the sample / mean fluorescence value of the standard). Plants were sampled randomly, with two subsamples measured per plant. Monoploid 1Cx genome size (i.e., the DNA content of one complete set of chromosomes) was calculated as 2C genome size / ploidy.
Chromosome counts were completed for selected taxa to confirm ploidy and further calibrate the flow cytometry results following the methods of Lattier et al. (2014). Root squashes were prepared for selected plants by collecting actively growing root tips and placing them in a prefixative solution of 2.0 mm 8-hydroxyquinoline plus 70 mg·L−1 cyclohexamide in the dark at room temperature for 3 h. The roots were then placed in the dark at 4 °C for another 3 h. After washing with dH2O, the roots were transferred to a fixative solution of 1:3 propionic acid to 95% EtOH at room temperature overnight. The following morning, roots were transferred to a solution of 70% EtOH for long-term storage.
To prepare fixed samples for counting, the roots were moved to a hydrolysis solution of 1:3 12 M HCl to 95% EtOH for 60 to 90 s before being moved to a clean slide. Root tips were excised and moved to a final clean slide, and a drop of modified carbol fuschin stain was applied to the root tip (Carr and Walker, 1961; Kao, 1975). A coverslip was placed on the root tip and gently pressed with a pencil eraser to squash the tissue. A light microscope was used to count chromosomes in actively dividing cells and confirm ploidy (Lattier et al., 2014).
Data for monoploid genome sizes (1Cx) were subjected to a one-way analysis of variance (ANOVA) as a function of subfamily, genus, and species; Camellia was subjected to a one-way ANOVA as a function of subfamily, genus, species, subgenus, and section. Means were separated using Fisher’s least significant difference test (Proc GLM; SAS version 9.4; SAS Institute, Cary, NC).
Our compilation of literature regarding the cytogenetics of Theaceae spans nearly one century of research and includes published results for 7 genera, 160 species, and 202 cultivars (Table 1). Because cytological studies of Camellia span a broad range of time, many of these referenced studies used different taxonomic treatments dependent on the naming conventions at the time of publication. Without vouchered specimens, it is essentially impossible to verify exactly which species, according to modern taxonomic treatments, were used during these previous studies. Therefore, the names in Table 1 remain as they were reported in the original publications. As a result, there are numerous taxa represented here by duplicate names, such as C. assamica, which is now treated as a variety of C. sinensis in the Flora of China (Ming and Bartholomew, 2007). Discrepancies regarding base chromosome numbers continue to be resolved over time, with more recent studies supporting that Theeae, including Apterosperma, Camellia, Polyspora, and Pyrenaria (Tutcheria), has 1n = 1x = 15 (Kondo et al., 1991; Yang et al., 2000, 2003, 2004). Additional work regarding Stewartiae, including Stewartia, indicated a base chromosome number of 1n = 1x = 17 (Horiuchi and Oginuma, 2001), whereas Gordoniae, including Franklinia, Gordonia, and Schima, had 1n = 1x = 18 (Bostick, 1965; Oginuma et al., 1994). Numerous Camellia species have isoploid series, including C. caudata, costeri, crapnelliana, forrestii, grijsii, hiemalis, japonica, kissii, mairei, nanyoungensis, octopetala, oleifera, pitardii, reticulata, rubituberculata, salicifolia, saluenensis, sasanqua, sinensis, tsaii, yuhsienensis, and yungkiangensis. There have been occasional reports of anisoploids in C. assamica, irrawadiensis, japonica, rosaeflora, sasanqua, sinensis, vietnamensis, and some cultivars of the interspecific hybrid C. ×vernalis that may have resulted from unreduced gametes, interploid hybridization, or nonrecurrent apomixis (Ozias-Akins and van Dijk, 2007). Previous research summarized in Table 1 has emphasized the cytogenetic diversity within Theaceae and has aided in the understanding of relationships between members of Theaceae. The previous cytogenetic and cytometric data serve as an accessible reference for plant breeders, taxonomists, and others studying Theaceae.
Cytology was completed for representative Theaceae, including species of Camellia, Gordonia, Polyspora, Pyrenaria, Schima, and Stewartia (Fig. 1). Results documented Camellia azalea (2018-063) as 2n = 2x = 30, Camellia japonica ‘Dr. JC Raulston’ as 2n = 2x = 30, Camellia sinensis (2017-111) as 2n = 2x = 30, Camellia ×vernalis ‘Egao Corkscrew’ as 2n = 4x = 60, Gordonia lasianthus (2006-220) as 2n = 2x = 36, Polyspora chrysandra (2015-114) as 2n = 2x = 30, Schima superba (2018-009) as 2n = 2x = 36, Stewartia pseudocamellia (2018-111) as 2n = 2x = 34, and Pyrenaria spectabilis (2018-008) as 2n = 2x = 30. These results further substantiate the base chromosome numbers for these genera and provide additional direct standards to further calibrate ploidy with genome size.
Citation: HortScience horts 54, 7; 10.21273/HORTSCI13923-19
Flow cytometry was completed for a broad range of taxa for Theaceae, providing data regarding 2C holoploid genome size, 1Cx monoploid genome size, and estimated ploidy for 123 nonhybrid accessions (Table 2). Our study represents new data for many cultivars of C. japonica, C. sasanqua, C. sinensis, C. oleifera, C. rosthorniana, and C. hiemalis. Furthermore, the ploidy level of seven previously unreported species of Camellia, including C. amplexicaulis, C. chrysanthoides, C. cordifolia, C. cucphuongensis, C. flava, C. nanyongensis, and C. trichoclada, was found to be diploid. The majority of tested Camellia species exhibited 2C genome sizes consistent with previously reported ploidy, although there were many exceptions (Table 1). The accession of C. assimilis CGBG2 was found to be diploid, consistent with the results of Fukushima et al. (1966) and Kondo (1977). Regarding Camellia (Ming and Bartholomew, 2007), C. assimilis is synonymous with C. caudata, which has been reported both as diploid (Bezbaruah, 1971; Zhuang and Dong, 1984) and tetraploid (Gu et al., 1988b; 1989b; Gu and Sun, 1997), indicating the existence of a possible ploidy series. Camellia brevistyla, reported by Zhang and Min (1999) as diploid, was found to be tetraploid in this study, although the tested accession was received as C. puniceiflora, which, according to Ming and Bartholomew (2007), is synonymous with C. brevistyla var. brevistyla. Camellia grijsii has a reported ploidy series including diploids (Gu et al., 1988b; 1989b; Huang et al., 2013; Lu et al., 1993; Xiao et al., 1991), tetraploids (Huang and Hsu, 1987; Kondo et al., 1991), pentaploids (Huang and Hsu, 1987), and hexaploids (Huang and Hsu, 1987; Xiao et al., 1991). The surveyed accession of C. grijsii was diploid, and the accession of C. odorata syn. C. grijsii var. grijsii (Ming and Bartholomew, 2007) was hexaploid. Both accessions of C. yuhsienensis, which is also synonymous with C. grijsii var. grijsii, were hexaploid, although C. yuhsienensis has been reported as tetraploid (Zhuang and Dong, 1984), pentaploid (Zhuang and Dong, 1984), and hexaploid (Huang et al., 2013; Xiao et al., 1993; Zhuang and Dong, 1984). Both accessions of C. lutchuensis were diploid, which was in agreement with previous reports (Ackerman, 1971; Kondo, 1977; Kondo and Parks, 1979). However, one of the accessions of C. lutchuensis was received as C. transnokoensis, which is synonymous with C. lutchuensis var. lutchuensis (Ming and Bartholomew, 2007). Kondo (1977) reported C. transnokoensis as hexaploid. The surveyed accession of C. reticulata ‘Captain Rawes’, which was reported to be triploid by Patterson et al. (1950), was found to be hexaploid. Camellia azalea (=C. changii), a relatively newly discovered species with considerable breeding potential, was estimated by Huang et al. (2013) to be hexaploid, although the accession of this species surveyed in this study was confirmed to be diploid through flow cytometry and cytology. This result was further supported by the diploidy of C. ‘Wendzalea’, a hybrid of C. azalea and C. japonica (diploid). The confusion and complexity of Camellia nomenclature and variations in ploidy within species emphasize the need to collect and reference data regarding individual clones and accessions.
The other genera of Theaceae included in this study are muc h less commonly cultivated and have been less studied compared with Camellia. Ploidy levels of 24 accessions of six other genera were determined, including Franklinia, Gordonia, Schima, Stewartia, Polyspora, and Pyrenaria. The genus Polyspora (many species of which were previously included in Gordonia) (Yang et al. 2004) was found to have a ploidy series ranging from 2n = 2x = 30 to 2n = 8x = 120. The ploidy levels of four species of Schima, including S. argentea, S. khasiana, S. remotiserrata, and S. sinensis, have been reported as diploid for the first time. A triploid accession of Stewartia ovata included in this study represented the first polyploid report of this genus; however, this was possibly the result of an unreduced gamete.
Monoploid genome sizes (1Cx) varied considerably by subfamily (Table 3), with the Gordoniae having a mean of 0.84 pg, and the Stewartiae and Theeae having substantially larger values of 2.50 and 3.00 pg, respectively. The much larger 1Cx values of the Stewartiae and Theeae indicated that they underwent considerable genome expansion independent of increased ploidy levels as these lineages diverged. Genome expansion such as this can occur through amplification of noncoding repetitive DNA including retrotransposons (Leitch and Leitch, 2013). The biological impact of genome size variation is still being elucidated, but the speciation rate has been shown to be correlated with the rate of genomic evolution and genome size (Bromham et al. 2015; Puttick et al., 2015). Within the subfamily Theeae, Camellia also had a significantly higher mean genome size of 3.13 pg compared with 1.75 pg for Polyspora and 1.39 pg for Pyrenaria, indicating differential rates of genome expansion among these groups. Even within Camellia, there were significant differences in 1Cx values among subgenera, sections, and some species (range, 2.70–3.57 pg), indicating that the evolution of chromosomes and genome size has been particularly dynamic compared with sister lineages.
Genome size has been used to estimate ploidy of interspecific hybrids, and it has been particularly useful for validating interploid and intergeneric hybrids (Table 4). All noninterploid, interspecific Camellia hybrids had estimated ploidy levels that were consistent with their reported parentage. However, the genome size of one putative intergeneric hybrid, C. japonica × F. alatamaha (USNA 79387) (Ackerman and Williams, 1982), was inconsistent with the reported parentage. Genome sizes are considerably different for these two parents (2C = 5.78–7.11 pg for C. japonica and 2C = 1.62 pg for F. alatamaha), yet the putative hybrid was 2C = 5.94, effectively discounting hybridity. Many putative interspecific interploid hybrids also had genome sizes and estimated ploidy levels that were inconsistent with their reported parentage. For example, ‘Arctic Dawn’, ‘Fire ‘N’ Ice’, ‘Ice Follies’, ‘Pink Icicle’, ‘Red Fellow’, ‘Spring Cardinal’, and ‘Spring Circus’ are all putative hybrids between hexaploid and diploid taxa, yet they have genome sizes consistent with diploids, suggesting they are the result of pollen contamination, mislabeling, or apomixis. Similarly, ‘Spring Frill’ is a putative hybrid between a hexaploid and tetraploid, but the estimated ploidy is diploid, suggesting mislabeling. Other interploid hybrids such as ‘Scarlet Temptations’ and ‘Starry Pillar’ are crosses between hexaploid and diploid taxa, but they were pentaploid, suggesting they are the result of an unreduced gamete from the diploid parent, as has been documented for Camellia (Wendel, 1984).
Camellia ×vernalis has been documented as a group of interspecific hybrids between C. sasanqua and japonica that were originally represented by F1 ‘Gaisen’-type tetraploids found on Hirado Island in Japan 400 years ago (Tanaka, 1988a, 1988b; Tanaka et al., 1986, 2005; Uemoto et al., 1980). These hybrids are fertile and can produce progeny that may have three-times, four-times, five-times, or six-times the number of chromosomes, or they may be aneuploid, depending on the ploidy of the other parent, occurrence of unreduced gametes, or other meiotic irregularities (Tateishi et al., 2007). Camellia ×vernalis ‘Ginryu’, also known as its westernized name ‘Dawn’, was found to be a triploid. This cultivar most likely resulted from a ‘Gaisen’-type tetraploid C. ×vernalis backcrossed to a diploid C. japonica. Open-pollinated seedlings derived from C. ×vernalis ‘Egao’ and ‘Star-Above-Star’ (both tetraploids) included triploids (most likely crossed with diploids), tetraploids (most likely crossed with other tetraploids), and hexaploids (most likely unreduced gametes from both the tetraploid C. ×vernalis and a diploid). Two triploid C. ×vernalis, ‘Christmas Candy’ and ‘Ginryu’, produced seedlings that were tetraploid (most likely producing unreduced gametes and crossed with diploids). Interestingly, ‘Ginryu’ also produced seedlings (‘Starman’ and GrN 08-070) that were triploid or nearly triploid and may have resulted from either apomixis or aneuploid/2x gametes.
The extensive history of Camellia breeding and selection has produced tens of thousands of cultivars that now serve as potential parents and breeding lines. Considerable progress has been made in resolving the taxonomy, systematics, and cytogenetics of the genus, but challenges remain. The long history of Camellia cultivation, global exchange of historical varieties, cultivar names that often relate to the origin of the variety or a quality of the flower, and variable translations can cause considerable confusion. One such name, ‘Shishigashira’, has been attributed to several species and hybrids, including C. japonica, C. sasanqua, and C. hiemalis, which some believe to be a form of C. sasanqua (Jiang et al., 2012). ‘Hiryu’ is another name that has been associated with C. japonica and C. sasanqua, as well as with C. ×vernalis, which is the hybrid of those two species. This confusion is further complicated by incomplete knowledge of the parentage and ploidy, along with the potential for pollen contamination, mislabeling, and variable reproductive pathways (e.g., unreduced gametes, apomixis, etc.). These challenges underscore the need for clone-specific data regarding cytogenetics for individual accessions and breeding lines.
This study builds on an extensive body of cytogenetic research regarding Camellia and provides new information regarding ploidy, genome size, hybridity, and reproductive pathways for a broad range of cultivated Camellia and related genera. This expanding knowledgebase provides improved characterization of genetic resources for Theaceae that will aid in the development of improved hybrids and cultivars.
Photomicrographs of condensed stained chromosomes of Theaceae. (A) Gordonia lasianthus 2006-220, 2n 2x = 36. (B) Schima superba 2018-009, 2n = 2x = 36. (C) Stewartia pseudocamellia 2018-111, 2n = 2x = 34. (D) Camellia japonica ‘Dr. JC Raulston’ 2017-060, 2n = 2x = 30. (E) Camellia sinensis ‘Red Leaf’ 2017-111, 2n = 2x = 30. (F) Camellia azalea 2018-063, 2n = 2x = 30. (G) Camellia ×vernalis ‘Egao Corkscrew’ 2017-062, 2n = 4x = 60. (H) Polyspora chrysandra 2015-114, 2n = 2x = 30. (I) Pyrenaria spectabilis 2018-008 2n = 2x = 30.
Contributor Notes
This research was funded, in part, by the North Carolina Agricultural Research Service (NCARS), Raleigh, NC. Use of trade names in this publication does not imply endorsement by the NCARS of products named or criticism of similar ones not mentioned. We gratefully acknowledge Clifford and David Parks, Camellia Forest Nursery, Chapel Hill, NC; Gene Phillips, Gene’s Nursery, Savannah, GA; Coastal Georgia Botanical Garden, Savannah, GA; Bobby Green, Green Nurseries, Fairhope, AL; Ray Watson, CAM TOO Camellia Nursery, Inc., Greensboro, NC; Scott McMahan, Atlanta Botanical Garden, Atlanta, GA; Gary Knox, University of Florida, North Florida Research and Education Center, Quincy, FL; Kevin Conrad, United States National Arboretum, Washington, D.C.; Donglin Zhang, University of Georgia, Athens, GA; John Ruter, University of Georgia, Athens, GA; Jack Johnston, Lakemont, GA; Star Roses and Plants, West Grove, PA; Gulf Coast Camellia Society, Baton Rouge, LA; Nathan Lynch, Irene Palmer, and Andra Nus, Mountain Crop Improvement Lab, Mills River, NC; and the staff at the Mountain Horticultural Crops Research and Extension Center for their cooperation, technical assistance, plant material, and/or funding for this project.
Graduate Research Assistant.
JC Raulston Distinguished Professor.
Associate Professor.
Director, JC Raulston Arboretum.
Corresponding author. E-mail: tom_ranney@ncsu.edu.
Photomicrographs of condensed stained chromosomes of Theaceae. (A) Gordonia lasianthus 2006-220, 2n 2x = 36. (B) Schima superba 2018-009, 2n = 2x = 36. (C) Stewartia pseudocamellia 2018-111, 2n = 2x = 34. (D) Camellia japonica ‘Dr. JC Raulston’ 2017-060, 2n = 2x = 30. (E) Camellia sinensis ‘Red Leaf’ 2017-111, 2n = 2x = 30. (F) Camellia azalea 2018-063, 2n = 2x = 30. (G) Camellia ×vernalis ‘Egao Corkscrew’ 2017-062, 2n = 4x = 60. (H) Polyspora chrysandra 2015-114, 2n = 2x = 30. (I) Pyrenaria spectabilis 2018-008 2n = 2x = 30.
