Camellia sasanqua botany

Yuri Panchul, June 2003



There are three most recent classification systems of the genus Camellia frequently referred in Camellia literature: Sealy 1958 [4], Chang 1981 [1] and Ming 2000 [3].

Taxonomy – Sealy

J. Robert Sealy divided genus Camellia into 12 species group (sections). He put C. sasanqua, C. oleifera and C. kissi into section Paracamellia, C. hiemalis and C. miyagii into unplaced (artificial) section Dubiae.

Sealy’s Paracamellia consisted of ten species. Their main feature was short styles and minimal fusion of floral parts.

In 1971 Dr. William L. Ackerman shown in his article [5] that C. hiemalis and C. miyagii freely hybridize with species of section Paracamellia and suggested they should be in one section.

Taxonomy – Chang

Chang Hung Ta (1981, [1]) divided genus Camellia into four subgenera and 20 sections. He put C. sasanqua and C. oleifera into section Oleifera of Camellia subgenus. Then he put C. kissi and C. miyagii into section Paracamellia of the same subgenus and C. hiemalis into section Camellia subsection Reticulata of the same genus.

We believe later Chang Hung Ta corrected C. hiemalis classification and put it back into section Paracamellia.

Chang stated that the reason five species should be put into a separate Oleifera section is because they have more stamen series and relatively longer styles. Xiao Tiaojiang and Clifford Parks (2002, [10]) doubted Chang’s reasons for dividing Paracamellia into two sections (Paracamellia and Oleifera). They noticed that wild forms of C. sasanqua (Changs’s section Oleifera) and C. miyagii (Chang’s section Paracamellia) are virtually identical and can be considered a one species. They also did DNA sequence analysis and found all species of Changs’s Oleifera section to be clustered with a group of species in Paracamellia section.

Xiao Tiaojiang and Clifford Parks also shown by DNA analysis that some of Chang’s Paracamellia species may be in fact not belonging to Paracamellia section, for example C. grijsii, C. odorata and C. yusienensis. They fall into the clade of section Camellia species from Western China.

Taxonomy – Ming

Ming Tianlu (2000, [3]) divided genus Camellia into two subgenera and 14 sections. We do not have his book so we cannot describe his treatment of Paracamellia species. Neither Sealy nor Chang recognized C. vernalis to be a separate species. In fact, many researchers consider C. vernalis to be a complicated sasanqua-japonica hybrid (see the details below). Some researches also consider C. hiemalis a sasanqua-japonica hybrid.

According to William Ackerman, when he traveled in 1980 on a plant exploration trip to western Japan, he saw wild populations of both C. sasanqua and C. japonica growing adjacent to each other, and intermingled. There were also obvious hybrids showing intermediate phenotypic characteristics. Ackerman’s cytological analysis of a series of C. vernalis cultivars showed chromosomal evidence of both 1st and 2nd generation hybridization.

On the other hand, Ackerman strongly disagree with those who consider C. hiemalis a hybrid with C. japonica parentage. He does not see neither cytological nor phenotypical evidence to support this.


Species by Chang Hung Ta classification

Section Oleifera Chang

C. gauchowensis Chang (1961)
C. lanceoleosa
C. oleifera Abel (1818)
C. sasanqua Thunb. (1784)
C. vietnamensis Hung ex Hu (1965)

Section Paracamellia Sealy

C. brevistyla (Hay.) Cohen-Stuart (1908)
C. confusa (Craib) Cohen-Stuart (1916)
C. fluviatilis Hand.-Mazz. (1922). (Synonim C. kissi)
C. grijsii Hance (1879)
C. hiemalis Nakai (1940)
C. maliflolia Lindl. (1827)
C. microphylla (Merr.) Chien (1937)
C. miyagii (Koidz.) Mak. & Nem. (1931)
C. obtusifolia Chang (1981)
C. odorata
C. phaeoclada Chang (1981)
C. puniceiflora Chang (1981)
C. shensiensis Chang ex Chang (1981)
C. tenii Sealy (1949)
C. weiningensis Y.K. Li ex Chang (1981)
C. yuhsienensis Hu (1965)

Section Paracamellia Sealy – not in Chang’s list, but from the International Camellia Society website:

C. brevissima Chang & Liang (1982)
C. lutescens Dyer in Hook. (1874)
C. octopetala Hu in Acta Phytotax. Sin. vol.X, No.2, 1965
C. paucipetala Chang, (1984).


According to William L. Ackerman (1971, [5]), C. sasanqua, C. oleifera and C. kissi of Sealy’s section Paracamellia hybridize with each other very readily. In Ackerman’s research the compatibility ratio of hybrids in relation to total cross-polunations was 29 percent, the highest withing any of the section he experimented.

Ackerman also hybridized hiemalis and C. miyagii of Sealy’s section Dubiae (Chang’s section Paracamellia). The compatibility ratio was 19 percent.

Ackerman also found that C. hiemalis and C. miyagii of Sealy’s section Dubiae hybridized as easily as when intrasectional crosses were made within Sealy’s section Paracamellia (C. sasanqua, C. oleifera and C. kissi). The compatibility ratio was 18 percent for C. miyagii and 13 percent for C. hiemalis.

All these percentage numbers compare with just 9 percent for intrasectional crosses within section Camellia.

Ackerman indicated that C. sasanqua, C. oleifera and C. kissi are ecospecies. He also suggested C. hiemalis and C. miyagii are ecospecies as well and should be put into Sealy’s section Paracamellia.

In Ackerman’s experiments section Thea appeared to be more closely related to section Paracamellia and to C. hiemalis and C. miyagii of Dubiae than to species of other sections.


The basic chromosome number in the genus Camellia is 15. Different species have chromosome numbers of 30, 45, 60, 75 and 90. According to Ackerman [5] C. sasanqua, C. hiemalis, C. oleifera and C. miyagii are generally hexaploids (chromosome number 6X=90).

C. kissi is a diploid (2X=30).

C. sasanqua ‘Narumigata’ is a pentaploid (5X=75)

C. vernalis ‘Hiryu’ was reported to be a triploid (3X=45) by Longley and Tourje (1959 [6], 1960 [7]).

Most C. japonica and C. sinensis are diploid (2X=30).

There are rare cases of triploid C. sinensis (3X=45).

The following numbers of chromosomes were reported by Ackerman [5] for crosses:

C. japonica 30 x C. kissi 30 = 30
C. kissi 30 x C. rusticana 30 = 30
C. japonica 30 x C. miyagii 90 = 60
C. sasanqua ‘Narumigata’ 75 x C. granthamiana 60 = 60
C. sasanqua ‘Narumigata’ 75 x C. reticulata 90 = 90
C. oleifera 90 x C. hiemalis 90 = 90
C. oleifera 90 x C. miyagii 90 = 90
C. reticulata 90 x C. sasanqua 90 = 90
C. sasanqua 90 x C. hiemalis 90 = 90
C. sasanqua 90 x C. miyagii 90 = 90
C. sasanqua 90 x C. miyagii 90 = 86
C. sasanqua 90 x C. oleifera 90 = 90
C. sasanqua 90 x C. reticulata 90 = 90

According to Ackerman [5] “‘Narumigata’, a pentaploid variety of C. sasanqua, produced hybrids when used as the female parent. However, the chromosome number of its hybrids seem unpredictable. A hybrid, A-24, resulting from C. sasanqua ‘Narumigata’ (5X=75) x C. granthamiana (4X=60) was tetraploid (4X=60). The morphological characters of this hybrid were intermediate. It is generally difficult to assess accurately the contribution of each parent to the hybrid in crosses involving polyploid species without the aid of genetical or cytological markers. However, ‘Narumigata’ may have produced an egg with 30 chromosomes, which united with a sperm carrying 30 chromosomes from C. granthamiana. A hybrid of C. sasanqua ‘Narumigata’ x C. reticulata (6X=90) was hexaploid. In this case, ‘Narumigata’ may have produced an egg cell with 45 chromosomes.”

William Ackerman also reports in his recent correspondence C. vernalis tetraploid (4X=60) and pentaploid (5X=75). This is what one would expect along the following lines, which substantiates the hybrid nature of C. vernalis:

  • 1st Generation (F1) hybrid between C. sasanqua 6X=90 x C. japonica 2X=30 with result in gametes 45 + 15 = 60 chromosomes (4X,tetraploid).
  • Backcross of resulting F1 hybrid to C. sasanqua: F1 hybrid 4X=60 x C. sasanqua 6X=90 will result in gametes 30 + 45 = 75chromosomes (5X, pentaploid).
  • Backcross of resulting F1 hybrid to C. japonica: F1 hybrid 4X=60 x C. japonica 2X=30 will result in gametes 30 + 15 = 45chromosomes (3X, triploid). This triploid will normally be sterile.


[1] Chang Hung Ta. 1981. A taxonomy of the genus Camellia. In Chinese. Acta Scientarum Naturalium Universitatis, Sunyatseni

Chang’s book was revised in 1998 (also in Chinese). English translation of 1981 Chang’s book is available on

[2] Chang Hung Ta, Bruce Bartholomew. 1984. Camellias. Timber Press, Portland, Oregon.

[3] Ming Tianlu. 2000. Monograph of the genus Camellia. Yunnan Science and Technology Press, Kunming, P.R. China

[4] J. Robert Sealy. 1958. A Revision of the Genus Camellia. The Royal Horticultural Society, London

It is possible to buy Sealy’s book on the Internet


[5] William L. Ackerman. 1971. Genetic and cytological studies with Camellia and related genera. Washington, D. C.

[6] Longley, A. E., and Tourje, E. C. Chromosome numbers of certain camellia species and allied genera. American Camellia Yearbook. 1959: 33-39.

[7] Longley, A. E., and Tourje, E. C. Chromosome numbers of certain camellia species and allied genera. American Camellia Yearbook. 1960: 70-72.

[8] Clifford Parks, K. Kondo and T.Swain. Phytochemical evidence for the genetic contamination of Camellia sasanqua Thunberg. Japanese Journal of Breeding 31(2):168

[9] John M. Ruter. Nursery production of Tea Oil Camellia under different light levels. Trends in new crops and new uses. 2002. J. Janick and A. Whipkey (eds.). ASHS Press, Alexandria, VA.

[10] Xiao Tiaojiang, Clifford Parks. 2002. Molecular analysis of the genus Camellia. University of North Carolina, USA.

1 Response to “Camellia sasanqua botany”

Comments are currently closed.