®
Floriculture and Ornamental Biotechnology ©2012 Global Science Books
Cytogenetic and Phylogenetic Review of the Genus Lachenalia
Riana Kleynhans1,2* • Paula Spies2 • Johan J. Spies2
1 Agricultural Research Council (ARC), Roodeplaat Vegetable and Ornamental Plant Institute (VOPI), Private Bag X293, Pretoria 0001, South Africa
2 Department of Genetics (116), University of the Free State, P.O. Box 339, Bloemfontein 9300, South Africa
Corresponding author: * Rkleynhans@arc.agric.za
ABSTRACT
The genus Lachenalia (family Asparagaceae), endemic to southern Africa, is a horticultural diverse genus, with many species featuring in
the red data list of southern Africa. The extensive morphological variation within some species complicates species delimitation and has
led to taxonomic confusion. The genus is utilised in a breeding programme where cytogenetic and phylogenetic information is important
for the development of breeding strategies. Chromosome numbers of 89 species have been recorded in literature, with 2n = 10 to 56 and n
= 5 to 28. B-chromosomes have been described in some species. Basic chromosome numbers include x = 5, 6, 7, 8, 9, (probably 10), 11,
(probably 12), 13 and (probably 15). Polyploidy was reported in 19 taxa (23%), and is most common in the x = 7 group. Molecular
cytogenetic studies using 5S rDNA, 18S rDNA probes and DAPI staining, as well as molecular systematic studies using trnL-F and ITS12 were used to assess the phylogeny of the genus. All these studies indicated that species with the same basic chromosome number are
closely related. The one deviation is that it appears as if there are two separate groups within the x = 7 group. The cytogenetic and
molecular studies are further supported by breeding studies, where improved results are generally obtained from crosses within a
phylogenetic group or between closely related groups. This review of the literature reveals how different studies obtain similar results
regarding the phylogenetic relationships within the genus and how these results can be utilized to improve breeding strategies. It also
accentuates that further multidisciplinary studies are needed to solve the evolutionary history of the complex genus Lachenalia.
_____________________________________________________________________________________________________________
Keywords: chromosome numbers, cladograms, cross-ability, phylogeny, polyploidy
Abbreviations: APG, Angiosperm Phylogeny Group; atpB, ATPase beta chain; DAPI, 4',6-diamidino-2-phenylindole; FISH, Fluorescent in situ hybridization; ITS1-2, Internal transcribed spacer 1 and 2; MEGA, Molecular Evolutionary Genetics Analysis; n, gametic
chromosome number; RAPD, Random amplified polymorphic DNA; rbcL ribulose bisophosphate carboxylase (large); SANBI, South
African National Biodiversity Institute; trnL, leucyl-transfer RNA intron; trnF, phenylalanine-transfer RNA; VOPI, Vegetable and Ornamental Plant Institute; x, basic chromosome number; 2n, somatic chromosome number; 5S rDNA and 18S rDNA, 5S and 18S ribosomal
DNA
CONTENTS
INTRODUCTION........................................................................................................................................................................................ 98
CYTOGENETIC STUDIES......................................................................................................................................................................... 99
Chromosome counts ................................................................................................................................................................................ 99
Chromosome morphology ..................................................................................................................................................................... 101
Basic chromosome numbers and polyploidy ......................................................................................................................................... 105
Meiotic studies....................................................................................................................................................................................... 106
PHYLOGENETIC STUDIES .................................................................................................................................................................... 106
The phylogenetic position of Lachenalia .............................................................................................................................................. 106
Phylogeny within the genus................................................................................................................................................................... 107
CROSS-ABILITY IN LACHENALIA ........................................................................................................................................................ 107
COMPARISON BETWEEN CROSS-ABILITY, CYTOGENETIC AND MOLECULAR DATA ............................................................ 108
Basic chromosome numbers and cladograms ........................................................................................................................................ 108
Basic chromosome number and cross-ability ........................................................................................................................................ 108
Evolution and relatedness of different basic chromosome numbers ...................................................................................................... 109
Existence of basic chromosome numbers .............................................................................................................................................. 112
Existence of hybrid species ................................................................................................................................................................... 113
CONCLUSION .......................................................................................................................................................................................... 113
ACKNOWLEDGEMENTS ....................................................................................................................................................................... 114
REFERENCES........................................................................................................................................................................................... 114
_____________________________________________________________________________________________________________
INTRODUCTION
The genus Lachenalia Jacq. f. ex Murray, previously a
member of the family Hyacinthaceae (Manning et al. 2004;
Duncan and Edwards 2006, 2007), but since 2009 reclassified under the family Asparagaceae Juss. (APG III group
2009), is endemic to southern Africa. The genus now also
Received: 16 December, 2010. Accepted: 20 December, 2012.
includes the former genus Polyxena (Manning et al. 2004).
Lachenalia is a horticultural diverse genus, with a distribution range extending from the south-western coast of Namibia, southward throughout the Northern, Western and Eastern Cape provinces of South Africa (Duncan 1998). One
species extends as far inland as the south western part of the
Free State Province (Duncan 1996). Of the 126 species and
Invited Review
Floriculture and Ornamental Biotechnology 6 (Special Issue 1), 98-115 ©2012 Global Science Books
Fig. 1 Morphological variation in Lachenalia in the greenhouse.
and appearance, collectors have recognized the horticultural
potential of the genus for centuries (Duncan 1988; Du
Plessis and Duncan 1989; Kleynhans 2009, 2011; Reinten
et al. 2012). The huge phenotypic variation was also the
most important reason for the initiation of a breeding programme at the Agricultural Research Council in South Africa.
This led to the production of various hybrids and the introduction of new products to the international pot plant market (Fig. 3) (Kleynhans 2006).
The variability of the genus in terms of morphology and
cytogenetics, however, lead to specific challenges for the
breeding of new cultivars. Both incompatibility and other
isolation barriers exists (Kleynhans and Hancke 2002). A
large number of inter-species crosses are unsuccessful
(Kleynhans et al. 2009) and future breeding progress is
dependent on information about the genetic variation in the
genus. Results generated from cytogenetic and phylogenetic
research has value for the breeding programme (Kleynhans
et al. 2009) and can furthermore assist in the classification
and delimitation of species (Crosby 1986; Spies et al. 2002).
This paper reviews the current information available on
cytogenetics and phylogeny for the genus Lachenalia and
correlates this information to breeding results on crossability with the aim to draw some conclusions on relationships among the different species within the genus.
subspecies described, 10% are endangered, 17% are vulnerable, 2% are considered to be near threatened, 6% are
critically rare, 9% are rare and 2% are declining (SANBI
2009).
The genus is geophytic, deciduous and is usually winter
growing. The centre of diversity is in the Worcester grid
(3319) in the Western Cape province of South Africa, with
species diversity decreasing toward the eastern and northern
parts of its range (Duncan 2005). Although Lachenalia species like L. bulbifera and L. obscura are widely distributed,
a substantial number of species (e.g. L. moniliformis, L.
mathewsii) have a restricted distribution, contributing to the
vulnerability of these species (Duncan 1998).
Lachenalia occurs in a wide range of habitats, ranging
from arid to high rainfall areas. Lachenalia rubida for
example always grows in deep, pure sand often very close
to the sea, whilst a species like L. campanulata on the other
hand is found in heavy soil at altitudes exceeding 2000
metres (Duncan 1998). Between these two extremes, there
is a multitude of other habitats, including humus-rich soil
on granite, mineral rich soil, barren stony flats, limestone
outcrops and seasonally inundated, heavy clays (Duncan
1998).
The morphological diversity within the genus is well
known (Fig. 1). Variation occur in several morphological
characters, such as plant size, leaf number and posture,
flower-size, -colour and -orientation and flowering period
(Fig. 2). The extensive morphological variation within
some species complicates species delimitation and has led
to considerable taxonomic confusion (Duncan 1992). Several attempts have thus been made to establish some subgeneric classification within this complex genus, starting
with the work by Baker (1897), who divided the genus into
five sub-genera based on morphology. The first cytogenetic
work by Moffett (1936), however, already indicated that
true relationships cut across the groups of Baker and this
has been confirmed by various studies (Crosby 1986; Spies
2004; Hamatani et al. 2009; amongst others).
Due to the extensive morphological diversity in colour
CYTOGENETIC STUDIES
Chromosome counts
Lachenalia is unusually variable in chromosome number
with the presence of different basic chromosome numbers
(Moffett 1936; Crosby 1986; Johnson and Brandham 1997),
polyploidy (Kleynhans and Spies 1999) and B-chromosomes (Hancke and Liebenberg 1990; Johnson and Brandham 1997). The first cytogenetic studies on the genus came
from Moffett (1936). Chromosome numbers steadily increased over many years with information coming from
various authors (Table 1). Currently the chromosome num99
Review of the genus Lachenalia. Kleynhans et al.
A
B
C
D
E
F
H
G
Fig. 2 Morphological variation in different Lachenalia species. (A) L. aloides; (B) L. carnosa; (C) L. splendida; (D) L. bulbifera; (E) L. longibracteata; (F) L. violacea; (G) L. contaminata; (H) L. pustulata.
bers of 89 species have been recorded in literature. Somatic
chromosome numbers vary from 10 to 56 and gametic numbers from 5 to 28.
The cytogenetics is further complicated by varying
chromosome number reports for a number of species (Table
1). Deviating chromosome counts can first of all be explained by suspected wrong identification of species. In the
species L. orchioides the variation could most probably be
ascribed to accessions being wrongly identified. Crosby
(1986) reported that he received both L. fistulosa and L.
pustulata under the name of L. orchioides. Schlechter also
identified an accession of L. pallida as L. orchioides (Barker 1983). Both L. pallida and L. pustulata have chromosome numbers of 2n = 16 which could explain some of the
variation reported for L. orchioides. Lachenalia contaminata similarly has both 2n = 14 and 2n = 16 reported in
literature (Table 1). Gouws (1965) was the first to report
both these numbers. The author, however, described these
two numbers in one specific bulb of L. contaminata exhibiting cells with both 2n = 14 and 2n = 16. In this case the 2n
= 16 could be B-chromosomes that was not identified. Most
other chromosome counts of this species, except two by
Spies et al. (2008, 2009), are 2n = 16. In this species the
variation is not a case of mistaken identity and further investigation is needed to explain the variation.
The small size of the chromosomes (Hancke and Liebenberg 1990; Spies et al. 2000) in the genus can furthermore contribute to miscounts and possible miss-identification of B-chromosomes. The presence of B-chromosomes in
Lachenalia was described by Hancke and Liebenberg
(1990). According to the authors, B-chromosomes in Lachenalia do not have a specific staining pattern and are similar
in size to the smallest chromosome in the normal complement. This behaviour makes them difficult to identify
and therefore could explain some erroneous counts, reported in literature. B-chromosomes in Lachenalia do not
occur in all cells of a specific individual and also not in all
plants of a specific accession (Hancke and Liebenberg
1990). It is thus important to investigate the chromosome
number of several individuals from a specific population to
have accurate chromosome counts and correctly identify the
presence of B-chromosomes. Counting insufficient number
of cells can similarly lead to miscounts due to chromosome
damage occurring during slide preparation.
B-chromosomes have been reported in eight species,
namely L. aloides, L. anguinea, L. bulbifera, L. carnosa, L.
contaminata, L. obscura, L. reflexa and L. splendida
(Crosby 1986; Hancke and Liebenberg 1990; Johnson and
Brandham 1997; Kleynhans and Spies 1999; Spies et al.
2009). Hamatani et al. (1998) also reported an expected B100
Floriculture and Ornamental Biotechnology 6 (Special Issue 1), 98-115 ©2012 Global Science Books
A
B
D
E
G
C
F
H
I
Fig. 3 Different Lachenalia cultivars developed at ARC - Roodeplaat VOPI. (A) ‘Rosabeth’; (B) ‘Aqua Lady’; (C) ‘Cherise’; (D) ‘Namakwa’; (E) L.
bulbifera x L. rubida; (F) L. unicolor x L. splendida; (G) ‘Romaud’; (H) ‘Rainbow Bells’; (I) L. bachmannii x L. carnosa.
chromosome in a 2n = 23 accession of L. zeyheri. Another
example where possible B-chromosomes have not been
identified, can be found in L. barkeriana where both 2n =
14 and 2n = 16 was reported (Table 1). The 2n = 16 was,
however, only found in one cell (Müller-Doblies et al.
1987) of an otherwise 2n = 14 accession and could most
possibly be ascribed to extra chromosomes.
Chromosome morphology
The chromosome morphology of Lachenalia has been
described in various reports (Moffett 1936; De Wet 1957;
Hamatani et al. 1998; Hancke and Liebenberg 1998;
Hancke et al. 2001; Hamatani et al. 2004, 2007, 2009,
2010). Both Moffett (1936) and Hamatani et al. (1998, 2004,
2007) attempted to group the species of the genus based on
101
chromosome length and basic chromosome number. The
groupings by Moffett (1936) and Hamatani et al. (1998)
agreed, except for the division of the first group of Moffett
into two separate groups by Hamatani et al. (1998). Further
studies by Hamatani et al. (2004, 2007) added four groups
based on chromosome numbers and varying numbers of
larger chromosomes within specific basic chromosome
numbers.
Idiograms presented by De Wet (1957) do not agree
with karyograms by Moffett (1936) or Hamatani et al.
(1998, 2004, 2007). Neither does it agree with idiograms
presented by Hancke et al. (1998, 2001) and Hamatani et al.
(2009). The idiogram for L. aloides presented by Hancke et
al. (2001) agrees with Moffet’s division, but differs from
the karyograms of Hamatani et al. (1998, 2004, 2007) in
having 6 longer chromosomes and not only 2 long chromo-
Review of the genus Lachenalia. Kleynhans et al.
Table 1 List of Lachenalia species with the somatic- and gametic chromosome numbers reported in literature. Number in brackets (#) indicates number
of accessions for which the specific somatic or meiotic number was reported. All numbers were reported in the table under the current accepted botanical
name. Aneuploidy and other abnormalities or specific detail around polyploidy are indicated with superscripts.
Species
Somatic no. Gametic
Reference
(#)
no. (#)
L. alba W.F. Barker ex G. D. Duncan 18 (1), 20
Johnson and Brandham 1997
(3), 20/40 (1)
L. algoensis Schönland
14 (4)
Crosby 1986; Hamatani et al. 2007; Spies et al. 2008, 2009
7 (1)
Ornduff and Watters 1978
21 (1)
Hancke 1991
L. aloides (L.f.) Engl.
14 (32)+0-1B
Moffett 1936; Therman 1956; De Wet 1957; Mogford 1978; Crosby 1986; Hancke and
Liebenberg 1990; Hancke 1991; Johnson and Brandham 1997; Hamatani et al. 1998,
2004, 2007; Spies et al. 2008; Hamatani et al. 2009; Spies et al. 2009
7 (6)
Hancke and Liebenberg 1998; Moffett 1936
Crosby 1986
15 (1)1
Moffett 1936; Crosby 1986
21 (2)1
28 (7)
Crosby 1986; Hancke and Liebenberg 1990; Hamatani et al. 1998; Spies et al. 2009;
Hamatani et al. 2010
14 (1)
Ornduff and Watters 1978
L. ameliae W.F. Barker
18 (2)
Johnson and Brandham 1997
L. anguinea Sweet
30 (1)+2B
Johnson and Brandham 1997
L. arbuthnothiae W.F. Barker
14 (6)
Crosby 1986; Johnson and Brandham 1997; Hamatani et al. 1998; Spies et al. 2008,
2009
7 (1)
Spies et al. 2009
L. attenuata W.F. Barker ex G.D.
14 (1)
Spies et al. 2009
Duncan
L. bachmannii Baker
16 (5)
De Wet 1957; Crosby 1986; Johnson and Brandham 1997; Hamatani et al. 2004
L. barkeriana U. Müller-Doblies et 14 (3)
Müller-Doblies et al. 1987
al.
16 (2)
Nordenstam 1982; Müller-Doblies et al. 1987
L. bolusii W.F. Barker
18 (1)
Spies et al. 2009
L. bowkeri Baker
16 (1)
Dold and Philipson 1998
L. bulbifera (Cyrillo) Engl.
14 (1)
Crosby 1986
28 (7)
Kleynhans and Spies 1999; Spies et al. 2009
14 (1)
Ornduff and Watters 1978
Moffett 1936c; Crosby 1986; Johnson and Brandham 1997; Hamatani et al. 1998;
42 (15)+0Kleynhans and Spies 1999; Spies et al. 2008
1B1
49 (1)
Kleynhans and Spies 1999
56 (5)
Crosby 1986; Johnson and Brandham 1997; Kleynhans and Spies 1999
L. capensis W.F. Barker
16 (1)
Hamatani et al. 1998
28 (2)
Johnson and Brandham 1997; Spies et al. 2008
L. carnosa Baker
16 (26)
Crosby 1986; Johnson and Brandham 1997; Hamatani et al. 1998; Du Preez et al. 2002;
Spies et al. 2008; Hamatani et al. 2009; Spies et al. 2009
8 (1)+0-2B Spies et al. 2009
L. cernua G.D. Duncan
28 (1)
Spies et al. 2008
L. comptonii W.F. Barker
20 (5)
Crosby 1986; Johnson and Brandham 1997; Spies et al. 2009
10 (1)
Spies 2004
c26 (1)
Crosby 1986
L. concordiana Schltr. Ex W.F.
14 (1)
Spies et al. 2008
Barker
L. congesta W.F. Barker
26, 28 (1)
Johnson and Brandham 1997
L. contaminata Aiton
14 (3)
Gouws 1965; Spies et al. 2008, 2009
16 (11)+1B
De Wet 1957; Gouws 1965; Crosby 1986; Hancke 1991; Johnson and Brandham 1997;
Hamatani et al. 2004
8 (2)
Ornduff and Watters 1978
32 (1)
Johnson and Brandham 1997
L. convallarioides Baker
30 (1)
Johnson and Brandham 1997
L. doleritica G.D. Duncan
18 (2)
Spies et al. 2008, 2009
L. duncanii W.F. Barker
18 (1)
Spies et al. 2008
L. elegans W.F. Barker
14 (6)
Moffett 1936; Johnson and Brandham 1997; Spies et al. 2009
28 (12)
Moffett 1936; Crosby 1986; Johnson and Brandham 1997; Spies et al. 2009
14 (9)
Ornduff and Watters 1978; Spies et al. 2009
42 (4)
Johnson and Brandham 1997; Duncan 2001
21 (2)
Spies et al. 2009
De Wet 1957
56 (1)
28 (2)
Ornduff and Watters 1978
L. ensifolia (Thunb.) J.C. Manning
24 (3)
Johnson and Brandham 1997
and Goldblatt
26 (2)
Johnson and Brandham 1997; Hamatani et al. 2007
L. fistulosa Baker
14 (8)
Johnson and Brandham 1997; Spies et al. 2002; Hamatani et al. 2004; Spies et al. 2009
7 (2)
Ornduff and Watters 1978
28 (1)
Spies et al. 2008
L. framesii W.F. Barker
16 (3)
Du Preez et al. 2002; Spies et al. 2008
L. giessii W.F. Barker
32 (1)
Spies et al. 2008
L. gillettii W.F. Barker
16 (1)
Spies et al. 2008
102
Floriculture and Ornamental Biotechnology 6 (Special Issue 1), 98-115 ©2012 Global Science Books
Table 1 (Cont.)
Species
L. haarlemensis Fourc.
L. hirta (Thunb.) Thunb.
Somatic no.
(#)
18 (2)
Gametic
no. (#)
9 (1)
22 (6)
11 (2)
L. inconspicua G.D. Duncan
L. isopetala Jacq.
L. juncifolia Baker
24 (3)
18 (1)
30 (2)
40 (1)
22 (9)
11 (1)
L. karooica W.F. Barker ex G.D.
Duncan
L. klinghardtiana Dinter
L. kliprandensis W.F. Barker
L. lactosa G.D. Duncan
L. latimerae W.F. Barker
16 (1)
Reference
Johnson and Brandham 1997
Ornduff and Watters 1978
Johnson and Brandham 1997; Van Rooyen et al. 2002; Hamatani et al. 2004; Spies et
al. 2009
Ornduff and Watters 1978
De Wet 1957; Hancke 1991; Johnson and Brandham 1997
Spies et al. 2008
Johnson and Brandham 1997
Spies et al. 2008
Johnson and Brandham 1997; Hamatani et al. 2007; Spies et al. 2008, 2009; Hamatami
et al. 2010
Ornduff and Watters 1978
Duncan 1996
L. leomontana W.F. Barker
L. liliflora Jacq.
14 (2)
16 (1)
14 (1)
14 (1)
18 (2)
14 (1)
16 (7)
L. longibracteata Phillips
14 (4)
L. longituba (A.M. van der Merwe)
J.C. Manning and Goldblatt
L. macgregoriorum W.F. Barker
L. margaretae W.F. Barker
L. marginata W.F. Barker
28 (2)
Spies et al. 2008
Johnson and Brandham 1997
Spies et al. 2008
Spies et al. 2008
Hamatani et al. 2007, 2010
Spies et al. 2008
Moffett 1936; De Wet 1957; Hancke 1991; Johnson and Brandham 1997; Hamatani et
al. 1998, 2009; Spies et al. 2009
Moffett 1936
Crosby 1986; Hamatani et al. 2007; Spies et al. 2008; Hamatani et al. 2009
Ornduff and Watters 1978
Hamatani et al. 2007, 2010
22 (1)
14 (1)
14 (1)
28 (3)
29 (1)
10 (1)
Spies et al. 2008
Spies et al. 2008
Spies et al. 2008
Johnson and Brandham 1997
Johnson and Brandham 1997
Duncan 1996
14 (1)
Spies et al. 2008
26 (1)
14 (4)
16 (1)
Spies et al. 2008
Johnson and Brandham 1997; Hamatani et al. 1998; Spies et al. 2002, 2008, 2009
Spies et al. 2009
8 (1)
7 (2)
L. marginata subsp. neglegta Schltr.
Ex G.D. Duncan
L. marlothii W.F. Barker ex G.D.
Duncan
L. martinae W.F. Barker
L. mathewsii W.F. Barker
L. maximiliani Schltr. Ex W.F.
Barker
L. mediana Jacq.
14 (1)
18 (2)
26 (2)
9 (2)
13 (1)
L. minima W.F. Barker
L. moniliformis W.F. Barker
L. muirii W.F. Barker
L. mutabilis Sweet
18 (1)
22 (1)
14 (3)
10 (6)
5 (2)
12 (6)
6 (2)
14 (20)
7 (5)
L. namaquensis Schltr. Ex W.F.
Barker
24 (1)
56 (1)
16 (11)
8 (2)
L. namibiensis W.F. Barker
22 (2)
L. neilii W.F. Barker ex G.D. Duncan 18 (1)
L. nervosa Ker Gawll
16 (2)
8 (1)
L. obscura Schltr. Ex G.D. Duncan
L. orchioides (L.) Aiton
24 (2)
18 (2)+1B,
36 (2)
14 (20)
7 (19)
16 (5)
8 (1)
17 (1)1
Johnson and Brandham 1997
Spies et al. 2009
Crosby 1986; Spies et al. 2008
Spies et al. 2009
Spies et al. 2008
Spies et al. 2008
Johnson and Brandham 1997; Hamatani et al. 2007, 2009
Johnson and Brandham 1997
Ornduff and Watters 1978
Spies et al. 2000, 2009
Spies et al. 2002, 2009
De Wet 1957; Crosby 1986; Hancke and Liebenberg; 1990; Johnson and Brandham
1997; Hamatani et al. 1998; Spies et al. 2000, 2009
Hancke and Liebenberg 1998; Spies et al, 2002, 2009
Spies et al. 2000
De Wet 1957
Crosby 1986; Johnson and Brandham 1997; Du Preez et al. 2002; Hamatani et al. 2007;
Spies et al. 2008; Hamatani et al. 2009; Spies et al. 2009
Spies et al. 2009
Spies et al. 2008
Spies et al. 2008
Moffett 1936; Spies et al. 2008
Moffett 1936
Johnson and Brandham 1997; Hamatani et al. 2007
Johnson and Brandham 1997
Spies et al. 2008
Crosby 1986; Hamatani et al. 2007; Spies et al. 2008, 2009
Moffett 1936; Ornduff and Watters 1978; Spies et al. 2009
Moffett 1936; De Wet 1957; Hancke 1991
Moffett 1936
Moffett 1936
103
Review of the genus Lachenalia. Kleynhans et al.
Table 1 (Cont.)
Species
Somatic no.
(#)
18 (1)
28 (13)
Gametic
no. (#)
14 (2)
L. orthopetala Jacq.
L. pallida Aiton
24 (1)
29 (1)
16 (5)
16 (7)
8 (3)
L. patula Jacq.
L. paucifolia (W.F. Barker) J.C.
Manning and Goldblatt
L. peersii Marloth ex W.F. Barker
L. physocaulos W.F. Barker
L. polyphylla Baker
L. purpureo-caerulea Jacq.
16 (1)
26 (3)
L. pusilla Jacq.
14 (8)
L. pustulata Jacq.
16 (1)1
18 (1)
28 (1)
16 (24)
L. reflexa Thunb.
32 (1)1
14 (5)+0-2B
L. rosea Andrews
16 (1)
14 (6)
L. rubida Jacq.
21 (1)
28 (2)
14 (6)
L. splendida Diels.
28 (1)
16 (8)+2B
14 (3)
14 (1)
22 (1)
16 (4)
8 (2)
8 (2)
7 (1)
7 (1)
8 (2)
L. stayneri W.F. Barker
L. thomasiae W.F. Barker ex G. D.
Duncan
L. trichophylla Baker
18 (1)1
24 (1)
14 (1)
14 (2)
7 (1)
L. undulata Masson ex Bak.
L. unicolor Jacq.
20 (1)
16 (45)
8 (4)
L. unifolia Jacq.
32 (1)
16 (1)
21 (1)
22 (24)
11 (16)
L. valeriae G.D. Duncan
L. variegata W.F. Barker
L. ventricosa Schltr. ex W.F. Barker
L. verticillata W.F. Barker
L. violacea Jacq
24 (2)
26 (2)
44 (1)
16 (1)
14 (2)
12 (1)
28 (1)
14 (1)
16 (1)
14 (13)
7 (3)
15 (1)
16 (1)
Reference
Riley 1962
Moffett 1936; De Wet 1957; Crosby 1986; Johnson and Brandham 1997; Hamatani et
al. 2007; Spies et al. 2008; Hamatami et al. 2010
Moffett 1936; Ornduff and Watters 1978
Hancke and Liebenberg 1990
Johnson and Brandham 1997
Crosby 1986; Johnson and Brandham 1997; Spies et al. 2008, 2009
Moffett 1936; Crosby 1986; Johnson and Brandham 1997; Hamatani et al. 1998, 2004;
Spies et al. 2008, 2009
Moffett 1936; Ornduff and Watters 1978
Johnson and Brandham 1997
Johnson and Brandham 1997; Hamatani et al. 2007, 2010
Johnson and Brandham 1997; Hamatani et al. 2004; Spies et al. 2009
Spies et al. 2008
Spies et al. 2008
Moffett 1936; Johnson and Brandham 1997; Spies et al. 2009
Moffett 1936; Ornduff and Watters 1978
Crosby 1986; Müller-Doblies et al. 1987; Johnson and Brandham 1997; Hamatani et al.
1998, 2007, 2009
Nordenstam 1982
Spies et al. 2009
Hancke 1991
Moffett 1936; Crosby 1986; Johnson and Brandham 1997; Spies et al. 2000; Hamatani
et al. 2004; Spies et al. 2008
Moffett 1936; Ornduff and Watters 1978
Spies et al. 2000
Crosby 1986; Hancke and Liebenberg 1990; Johnson and Brandham 1997; Hamatani et
al. 1998; Spies et al. 2009
Hancke and Liebenberg 1998
De Wet 1957
Moffett 1936; Crosby 1986; Hancke 1991; Johnson and Brandham 1997; Hamatani et
al. 2007; Spies et al. 2008
Crosby 1986
Spies et al. 2009
Moffett 1936; Crosby 1986; Hamatani et al. 1998, 2009; Spies et al. 2009
Moffett 1936
Crosby 1986
Crosby 1986; Johnson and Brandham 1997; Hamatani et al. 1998; Du Preez et al. 2002;
Hamatani et al. 2009; Spies et al. 2009
Spies et al. 2009
Crosby 1986
Johnson and Brandham 1997
Spies et al. 2008
Johnson and Brandham 1997
Ornduff and Watters 1978
Johnson and Brandham 1997
Moffett 1936; De Wet 1957; Gouws 1965; Crosby 1986; Hancke 1991; Johnson and
Brandham 1997; Hamatani et al. 1998; Spies et al. 2000; Du Preez et al. 2002;
Hamatani et al. 2009
Moffett 1936; Ornduff and Watters 1978
Crosby 1986
Hancke 1991
De Wet 1957
Moffett 1936; De Wet 1957; Crosby 1986; Johnson and Brandham 1997; Van Rooyen et
al. 2002; Spies et al. 2009
Moffett 1936; Ornduff and Watters 1978; Spies et al. 2009
De Wet 1957; Hamatani et al. 2004
Moffett 1936; De Wet 1957
Johnson and Brandham 1997
Spies et al. 2008
Spies et al 2008; Hamatani et al. 2009
Hamatani et al. 2004
Spies et al. 2002
Spies et al. 2008
Crosby 1986
Hancke 1991; Johnson and Brandham 1997; Hamatani et al. 1998
Ornduff and Watters 1978; Spies et al. 2009
Johnson and Brandham 1997
Crosby 1986
104
Floriculture and Ornamental Biotechnology 6 (Special Issue 1), 98-115 ©2012 Global Science Books
Table 1 (Cont.)
Species
L. viridiflora W.F. Barker
Somatic no.
(#)
14 (7)
Gametic
no. (#)
7 (1)
L. youngii Baker
L. zebrina W.F. Barker
L. zeyheri Baker
16 (1)
30 (2)
22 (2)
23 (2)1
Reference
Nordenstan 1982; Crosby 1986; Hancke and Liebenberg 1990; Hancke 1991; Johnson
and Brandham 1997; Spies et al. 2002; Hamatani et al. 2007, 2009
Hancke and Liebenberg 1998
Spies et al. 2008
Johnson and Brandham 1997; Spies et al 2008
Johnson and Brandham 1997; Spies et al 2002
Hamatani et al. 1998, 2010
had 2n = 22 and L. unifolia as 27 out of 32 reports indicated
2n = 22 as somatic chromosome number);
x basic group x = 12 (L. ensifolia as 3 out of 5 reports
indicate 2n = 24 but this species can also be a possible x =
13 and L. stayneri because it formed a structural diploid
based on x = 12 rather than a tetraploid based on x = 6
(Johnson and Brandham 1997);
x three different basic chromosome numbers have been
recorded for L. mutabilis. This is the only species in basic
group x = 5, as well as basic group x = 6. The majority of
reports however comes from basic group x = 7 (24 out of
38).
Of the 83 taxa that could be grouped, basic x = 7 (41%)
and basic x = 8 (27%) were the most common, followed by
basic x = 9 (11%) and x = 11 (10%). Basic x = 10 (4%), x =
12 (2%), x = 13 (2%) and x = 15 (4%) are only present in a
small number of taxa (Table 1, Fig. 4). Basic x = 5 (1%)
and x = 6 (1%) were only present in L. mutabilis. Johnson
and Brandham (1997) stated that x = 5 reported for L. mutabilis were derived from plants with 2n = 14 via Robertsonian fusions. Based on their observations of no constant
number of long and short chromosomes in L. mutabilis,
Spies et al. (2000) disagreed with Johnson and Brandham’s
(1997) conclusion that the x = 5 L. mutabilis studied by
them resulted from Robertsonian fusions. Spies et al.
(2000) could not find any long chromosomes as a result of
Robertsonian fusions linked to specific specimens or a
specific basic number supporting the hypothesis of Johnson
and Brandham (1997). Spies et al. (2000) thus concluded
that the variation in L. mutabilis is more likely to be the
result of an aneuploid series. More studies are needed to
determine the actual mode of chromosome evolution in the
species L. mutabilis. Dysploid series also occurs in other
genera such as Prospero: x = 4, 5, 6, 7; Bernardia: x = 8, 9;
Hyacinthella: x = 9, 10, 11, 12 and Stellarioides: x = 2, 3, 4,
5, 6, 7, 8 and 9. Like in Lachenalia these aneuploid/dysploid series is difficult to interpret (Pfosser and Speta 1999).
Combining the chromosome counts with molecular and
morphological data might aid in the interpretation of the
chromosomal evolution in the genus.
The presence of polyploidy was reported in 19 Lachenalia taxa (23%), excluding L. capensis and L. congesta
where basic chromosome numbers could not be determined
from published results. Conclusions could thus also not be
drawn on polyploidy in these species (Table 1). Polyploidy
are most common in the basic x = 7 group, with 12 of the 34
species (35%) containing polyploid specimens and a few
species exhibiting a range of ploidy levels from triploid to
octoploid (Fig. 4; Table 1). Polyploidy were also reported
in basic group x = 6, 8, 9, 10 and 11, but here only tetraploids were observed. Tetraploids (present in 23% of the 83
grouped taxa) are the most common followed by octoploids
(4%) hexaploids (2%), triploids (2%) and heptaploids (1%).
Lachenalia bulbifera is the species with the largest
number of reported polyploid accessions including 4x, 6x,
7x and 8x accessions (Table 1). The heptaploid accession of
L. bulbifera originated from seed and it is thus possible that
the seed could have originated from an intra-species cross
between a 6x and an 8x individual (Kleynhans and Spies
1999). Specific ploidy levels in L. bulbifera were better correlated to geographic distribution than morphology (Kleynhans and Spies 1999).
somes. Idiograms for L. aloides and L. splendida constructed by Hamatani et al. (2009) again correlate with that of
Hancke et al. (2001).
Spies et al. (2000) reported that accessions of L. mutabilis contained 4 to 8 very short chromosomes. According
to the authors the number of short chromosomes can vary
between different localities and even between specimens
collected at the same locality. Hamatani et al. (2007) furthermore reported on varying karyotypes within the same
species for a number of Lachenalia species. This reported
variation and conflicting results thus indicate that karyomorphological data alone cannot be utilized successfully to
construct phylogenetic relationships in the genus Lachenalia. Similar conclusions were reached by Hamatani et al.
(2008), resulting in a movement towards molecular methods
to determine phylogenetic relationships in the genus.
Basic chromosome numbers and polyploidy
Moffett (1936) identified four different basic chromosome
numbers (x = 7, 8, 11 and 13) and polyploids, including 3x,
4x and 6x, in the x = 7 group. De Wet (1957) added a basic
chromosome number of x = 12 and reported on an accession with 2n = 56, a possible 8x. Ornduff and Watters
(1978) added x = 6, in an unidentified species as well as x =
5 and x = 9. Johnson and Brandham (1997) added x = 10
and 15.
For the purpose of this review, the 89 species in Table 1
was grouped according to their basic chromosome numbers.
Basic chromosome numbers of x = 5, 10 and 15 were also
included as existing basic numbers for the genus and not as
polyploid forms of basic group x = 5. Of the 89 species six
species (L. mediana, L. latimerae, L. isopetala, L. nervosa,
L. congesta and L. capensis) could not be placed into a specific basic chromosome number due to varying reports in
literature indicating different basic chromosome numbers
within these species. It is possible that L. mediana has two
different basic chromosome numbers and that x = 9 are present in L. mediana var. mediana and x = 13 are found in L.
mediana var. rogersii (Spies et al. 2008, 2009). More studies are, however, required to accurately place these six
species. Other species with varying chromosome number
reports were placed into specific groups according to the
most commonly reported chromosome number (Table 1).
These include:
x basic group x = 8 (L. contaminata 14 out of 17 reports
indicate 2n = 16);
x basic group x = 7 (L. barkeriana 3 out of four accessions had 2n = 14, L. marginata 4 out 5 reports indicate
either 2n = 14 or tetraploids of x = 7, L. orchioides – majority of reports indicate x = 7 and 2n = 16 most probably
from wrongly identified species, L. pusilla as 8 out of 9
reports indicate 2n = 14, L. reflexa as 5 out of 6 reports
indicate 2n = 14 and the 2n = 16 could most probably be
ascribed to the presence of B-chromosomes, L. variegata as
3 out 4 reports indicate basic x = 7 and L. violaceae as 15
out of 17 reports indicate basic x = 7);
x basic group x = 10 (L. alba as 4 out of 5 had 2n = 20
and Johnson and Brandham (1997) concluded that 2n = 20
forms a diploid based on x = 10 rather than a tetraploid
based on x = 5);
x basic group x = 11 (L. hirta as 8 out of the 12 reports
105
Review of the genus Lachenalia. Kleynhans et al.
35
30
25
2x
3x
20
4x
15
6x
7x
10
8x
5
0
x=5
x=6
x=7
x=8
x=9
x=10
x=11
x=12
x=13
x=15
Fig. 4 Basic chromosome numbers in the genus Lachenalia indicating the number of taxa for each basic number and the ploidy levels reported
for these basic numbers.
PHYLOGENETIC STUDIES
The only other species with ploidy levels above tetraploid are L. elegans and one report of 8x in L. mutabilis
(Table 1). The two triploid accessions in L. aloides and L.
rosea could have resulted from intra-species crosses
between diploid and tetraploid individuals in these species
followed by vegetative propagation or through an unreduced gamete followed by vegetative propagation as suggested by Moffett (1936).
Only a few molecular studies have been done on Lachenalia and most of these studies concentrated on the phylogenetic position of the genus. The extensive variation in the
genus, and even within a species, as indicated by RAPD
studies (Kleynhans and Spies 2000), complicates both the
phylogeny and taxonomy. In cultivation, a number of species are easily crossed and reproduce by means of offshoots
and bulb formation. The existence of possible natural hybrid species thus further complicates the phylogenetics of the
genus.
Meiotic studies
Reports on meiotic studies within the genus are less frequent. Moffett (1936) again presented the first report on
meiosis. The author found mostly normal meiosis for 2n =
14, 16 and 22 species. The only differences were reported
where ploidy was present. Hancke and Liebenberg (1998)
reported on the meiosis of several 2n = 14 species and
hybrids. Species studied displayed normal meiosis with 7
bivalents. Four of the six hybrids studied also displayed
normal meiosis with 7 bivalents indicting a close relationship between the species L. aloides, L. orchioides, L. viridiflora and L. reflexa. Two hybrids (both between L. aloides
and L. mutabilis) displayed a low percentage of trivalents
and quadrivalents. Hancke and Liebenberg (1998) presented
evidence of structural chromosomal changes involving
three chromosomes of which the acrocentric pair of
chromosomes was involved in at least one interchange. This
chromosome pair also seemed to be prominent in other
abnormalities observed during meiosis (Hancke and Liebenberg 1998).
Hancke et al. (2001) studied the chromosome associations of one interspecific dibasic hybrid between L. splendida and L. aloides and two interspecific dibasic hybrids
between L. unicolor and L. aloides. Results showed that L.
aloides is more closely related to both L. splendida and L.
unicolor than expected with genome affinity indexes of 0.9
and above. The results of the pairing configurations observed in these hybrids revealed homoeology between two
chromosomes of the x = 7 karyotype and three chromosomes of the x = 8 karyotype. This could indicate that the x
= 7 plants differ from the x = 8 plants by at least two exchanges of chromosome material and involves also the loss
of one centromere from the x = 8 karyotype. Hancke et al.
(2001) thus suggested that the change in basic chromosome
number of Lachenalia involves a reduction in number.
Du Preez et al. (2002) reported on normal meiosis with
8 bivalents for the following species, as well as the hybrids
between L. carnosa and L. splendida, L. splendida and L.
carnosa, L. unicolor and L. carnosa and L. carnosa and L.
framesii. This study indicated that these species are closely
related. Hamatani et al. (2009) confirmed this relationship.
The phylogenetic position of Lachenalia
The genus Lachenalia was included in several studies to
determine the phylogenetic position and classification of the
different species, the first being the inclusion of the genus
in the family Liliaceae. Lachenalia was reclassified in the
family Hyacinthaceae (Perry 1985) up to 2009, where after
the family Hyacinthaceae was dissolved into other families.
Lachenalia now belongs to the family Asparagaceae (APG
III group 2009).
To find the relative position of Lachenalia in the
Asparagaceae, Pfosser and Speta (1999) used sequences of
the trnL-F chloroplast region. From these results the authors
were able to group Lachenalia in the tribe Massonieae
(which consists of all the South African genera investigated,
such as Drimiopsis, Ledebouria and Polyxena). This study
also presented the first evidence suggesting a close relationship between Lachenalia and Polyxena, with a bootstrap
support of 100%. This was in contrast to that of MüllerDoblies and Müller-Doblies (1997), which placed Lachenalia in the subtribe Lachenaliinae and Polyxena into Massoniinae. Pfosser and Speta (1999) suggested further studies,
since only a few representative species were included in
their analysis.
A later study (Pfosser et al. 2003) included not only
more Lachenalia species, but also an additional chloroplast
region (atpB), as well as data on seed morphology. Polyxena, Lachenalia and the genus Periboea formed a monophyletic clade with a bootstrap support of 100%. This study
thus also supported the inclusion of Polyxena in the genus
Lachenalia. Within the monophyletic clade some species of
Lachenalia and Polyxena had low bootstrap support values
(66% and 62%, respectively) and it was suggested that the
specific delimitation may not be optimal for these clades.
Another explanation was that the species are more recently
derived, resulting in an insufficient number of base
substitutions to resolve the taxa. The authors suggested that
seed size and weight is higher in the basal genera such as
Eucomis, Merwilla and Ledebouria, with Veltheimia brac-
106
Floriculture and Ornamental Biotechnology 6 (Special Issue 1), 98-115 ©2012 Global Science Books
CROSS-ABILITY IN LACHENALIA
teata having seeds of 0.056 g and with a length of 6.1 mm.
The smallest seeds were found in the genus Lachenalia (L.
angelica: 0.0003 g; 0.9 mm long). Analysis on the seed size
and weight supports the hypothesis of the authors that
Lachenalia is a recently derived genus. The seed form and
structure of the micropylar swelling of the seed coat in
Lachenalia suggested that this genus was the most advanced in their study.
The inclusion of Polyxena in the genus Lachenalia was
raised again in three separate studies (Manning et al. 2004;
Spies 2004; Hamatani et al. 2008) using rbcL, trnL-F and
ITS1-2 sequencing data respectively. In all these studies,
Lachenalia and Polyxena formed a well supported monophyletic group. The two genera were characterised from
other genera in the family by their biseriate stamens with
the two series inserted at different heights. The two genera
can be distinguished from each other by the relative fusion
of the perianth (Manning et al. 2002). Manning et al. (2004)
thus included Polyxena within Lachenalia based on the
paraphyletic nature of the two genera.
Rev. John Nelson raised the first authenticated Lachenalia
hybrid in 1878 (Moore 1905). Since then a number of
claims of interspecific hybridization were published (Crosby 1978, for review of early work). None of these early
hybrids became available commercially. In 1965 the genus
was identified as an indigenous genus with potential for
development in South Africa. A breeding programme for the
development of flowering pot plants was started at the
Roodeplaat Vegetable and Ornamental Plant Institute of the
Agricultural Research Council and the first hybrids became
available commercially in 1997/1998 (Kleynhans 2006).
The extensive morphological and cytological variation
in the genus Lachenalia resulted in the existence of both
internal and external crossing barriers (Lubbinge 1980;
Kleynhans and Hancke 2002; Kleynhans 2006). External
crossing barriers like geographical separation and varying
flowering periods can be overcome through the cultivation
of species in controlled environments and the successful
storage of pollen for a 12 month period (Kleynhans 2006).
Internal crossing barriers include both post- and pre-fertilization barriers. Mechanical isolation (Lubbinge 1980) is
one of the first internal pre-fertilization barriers. Flower
length in Lachenalia species can vary from 5 to 30 mm
(Duncan 2005). Pollen from small flowered species is thus
not adapted to traverse the long distance from the stigma to
the ovary of large flowered species (Stebbins 1950). The
utilization of reciprocal crosses has been successful in overcoming this barrier (Lubbinge 1980; Kleynhans 2006).
Other pre- and post-fertilization barriers have not been studied in detail, but the extent of these barriers become clear
when the success rate of inter-species crosses are taken into
account.
For each crossing combinations at least 10 flowers,
within two different inflorescences were pollinated to ensure that wrong conclusions were not drawn, due to specific
physiological or developmental problems in the inflorescence or floret. Kleynhans et al. (2009) reported that only
33% of the inter-species crosses (1498) made over a 30 year
period were successful. With additional crosses (382) made
since 2005, this percentage dropped to only 18% (Table 2).
Of the 82% that did not succeed, 50% was related to the
absence of seed, indicating the presence of possible prefertilization barriers. A further 31% of the combinations
produced abnormal or non-viable seed that could be
ascribed to post-fertilization barriers. Lastly, 1% of the
crossing combinations did not succeed due to seedling
death shortly after germination. The reason for the death of
Phylogeny within the genus
Morphological studies have focused on the entire genus,
and many species have, over time, been included and
excluded and shifted around from one genus to another. The
first of these was when the genus was split into several
genera (Salisbury 1866). Later on the species in the genus
were sub-divided into smaller groups by Baker (1897),
Crosby (1986) and Duncan (1988, 2002). These groupings,
except for that of Crosby (1986) were based on different
morphological characteristics, and did not correspond with
each other.
Duncan et al. (2005) used morphological data of all the
species in the genus to construct a cladogram. The author
included 73 characters which comprised of 57 qualitative
and 16 quantitative characters. This study concluded that
Polyxena is paraphyletic with Lachenalia and forms the
basal clade. Many of the Lachenalia species formed polytomies or unrelated groups, but there were some synapomorphies or taxa sharing some traits.
Spies (2004) produced a cladogram based on chloroplast trnL-F sequencing data from 129 taxa, including four
Massonia taxa as outgroup. Hamatani et al. (2008) investigated nuclear ITS1-2 sequencing data of 56 taxa, including
two Massonia and one Ornithogalum as outgroup. Both
authors identified specific clades within the genus Lachenalia. The topologies of the cladograms produced by these
authors largely correspond.
Table 2 Number of inter-species crosses made among various different Lachenalia species over a 35 year period and the results obtained from these
crossing combinations. Crosses that did not succeed were linked to three different aspects namely no seed set, abnormal seeds or seedling death. Results
are linked to the basic chromosome complement of the species.
No of unsuccessful crosses
Basic chromosome number of parents
No. of successful
crosses
No. of crosses with no No. of crosses with
No. of crosses with
seed set
abnormal seed
seedling death
7x7
169 (27%)
274 (44%)
169 (27%)
10 (2%)
8x8
72 (46%)
44 (28%)
40 (45%)
1 (1%)
11x11
2 (67%)
1 (33%)
7x8
20 (6%)
251 (79%)
44 (14%)
3 (1%)
8x7
59 (18%)
111 (34%)
155 (47%)
6 (2%)
7x10
17 (100%)
10x7
1 (5%)
5 (25%)
13 (65%)
1 (5%)
7x11
1 (2%)
54 (86%)
8 (13%)
11x7
4 (6)
23 (33%)
39 (57%)
3 (4%)
9x8
1 (100%)
8x10
1 (33%)
2 (67%)
10x8
2 (33%)
1 (17%)
2 (33%)
1 (17%)
8x11
1 (3%)
23 (79%)
5 (17%)
11x8
1 (3%)
15 (39%)
22 (58%)
11x10
1 (100%)
15x7
2 (67%)
1 (33%)
Unknown basic numbers in one or both of the parents 4 (2%)
117 (59%)
78 (39%)
Total
336 (18%)
939 (50%)
580 (31%)
25 (1%)
107
Review of the genus Lachenalia. Kleynhans et al.
these seedlings can not necessarily be ascribed to hybrid
breakdown, as seedlings can also be affected by diseases.
The genetic variability within the genus as described
above has a direct influence on the cross-ability. With the
additional data presented in this review the comparison
between cross-ability and the cytogenetic and molecular
data will be discussed in the next section.
bers and phylogenetic groupings could in the future be used
to confirm basic numbers for species. A single count of 2n =
32 was reported for L. giessii but based upon a close phylogenetic grouping with x = 11 (Spies 2004), it seems that this
species could also be regarded as x = 11 (2n = 33) rather
than x = 8 (2n = 32). In this review it was included as a
tetraploid of x = 8 for the purpose of calculations, but this
species should be investigated further. Similarly L. capensis
groups with the x = 7 group (Spies 2004) thus supporting
the chromosome counts of Johnson and Brandham (1997)
and Spies et al. (2008) and suggesting that L. capensis
could be a basic x = 7 rather than a basic x = 8 as reported
by Hamatani et al. (1998). Further investigations and correct identification of species are, however, essential to solve
the inconsistent reports in chromosome numbers in some
species.
COMPARISON BETWEEN CROSS-ABILITY,
CYTOGENETIC AND MOLECULAR DATA
The complexity in the genus, in terms of morphology, cytogenetic and genetic variation complicates the determination
of the relationship within and between different species.
There are questions on the existence and origin of the
different basic chromosome numbers, as well as the mode
of speciation. Does the different basic chromosome numbers correlate with the phylogeny of the genus? Can the
phylogenetic information assist in the taxonomic grouping
of some difficult species and, furthermore, can phylogenetic
information shed some light on the existence of possible
natural hybrids? How does the phylogeny correlate with the
cross-ability between species and finally what conclusions
can be drawn when the different data sets are compared.
Basic chromosome number and cross-ability
Kleynhans et al. (2009) presented data showing that the
success rate of crossing combinations increased when
crosses were made between species containing the same
basic chromosome number. The information from additional
crosses made in the last five years were added to this data
and the number of successful crosses between species with
the same basic chromosome number was substantially
higher than between species from different basic chromosome numbers (Table 2). The success rate of crossing
combinations dropped to below 20% when species with
different basic chromosome numbers were crossed. The
only exception to this is the combination of basic x = 10
crossed with basic x = 8 (Table 2). The two successful
crosses resulted from a L. alba x L. unicolor and L. alba x L.
pustulata combination (specific results not shown).
The increased success rate reported between species
with the same basic chromosome number were a confirmation of a report by Crosby (1986) who also indicated that
species cross more readily within certain basic chromosome
number groupings. Based on differences in the cross-ability
and morphology the latter author also split the basic x = 7
group of species into two different groups. The existence of
different groupings within the basic x = 7 was confirmed by
Spies (2004) as discussed above. Meiotic data presented by
Hancke and Liebenberg (1998), as discussed above, also
indicated differences between especially the species L.
mutabilis and L. aloides as illustrated by structural chromosome changes. Kleynhans et al. (2009) used the three basic
clades as well as the phylogenetic groupings within the
basic x = 7 group as reported by Spies (2004) and presented
data that showed improved cross-ability when crosses were
made between individual species within the same phylogenetic groupings. The cross-ability was at least 10 to 20%
higher when crossing combinations were attempted within
the groups, than between groups. The cross-ability data thus
supported phylogenetic groupings as identified by Spies
(2004).
The close relationship illustrated in the phylogenetic
trees, between species with basic x = 8 was also confirmed
by the cross-ability data with a success rate of 46% (Table
2). The only success rate higher than this was that between
species with basic x = 11. This data, however, only included
3 crossing combinations in comparison to the 157 combinations within the basic x = 8 group and would most probably
decline with the inclusion of additional crossing combinations. The relationship among species with x = 8 was further
illustrated by Du Preez et al. (2002). In this meiotic study
several hybrids between different species with x = 8 were
investigated and all hybrids produced 8 bivalents. Hybrids
resulting from these crosses are also fertile and was successfully utilized in further crossing combinations (results
not shown).
Basic chromosome numbers and cladograms
A comparison between the groupings from Crosby (1986)
(based on chromosome numbers), Spies (2004) (chloroplast
trnL-F), Duncan (2005) (morphology) and Hamatani et al.
(2008) (nuclear ITS1-2) revealed that, with the exception of
a few species, there is a good correlation between the basic
chromosome numbers and the monophyletic groups identified in the different studies. When chromosome numbers
were superimposed on the cladogram of Duncan et al.
(2005) most of the x = 7 and x = 8 species fall into exclusive monophyletic groups for each chromosome number.
There are only two exceptions where x = 7 species (L. congesta and L. mathewsii) grouped with x = 8. Species with x
= 11 were closely related, even though they did not form a
monophyletic group. The rest of the chromosome numbers
form a polytomy. Although monophyletic groups linked to
basic chromosome numbers were obtained the morphological cladogram is poorly resolved for many of the species.
The study using trnL-F chloroplast DNA sequences
(Spies 2004) of 129 taxa distinguished several well defined
groups. The first group consisted of seven species with a
basic number of 11. Species with x = 7 and 8 formed a
monophyletic clade (the Lachenalia 1 group), suggesting a
close relationship between these two basic numbers. Within
this monophyletic clade, x = 8 formed a monophyletic subclade excluding only one species with a basic chromosome
number of x = 8, L. verticillata, and including L. pusilla (x
= 7), which was basal to this group. All species having a
basic chromosome number of x = 7, were distributed in different sister subclades, of which the two largest x = 7 subclades includes 25 and 10 taxa respectively. The second
largest group in the cladogram (the Lachenalia 2 group),
consisted of 48 poorly resolved taxa having chromosome
numbers of x = 6, 7, 8, 9, 10 and 13. This group has no
consistent pattern regarding chromosome numbers. These
results led the author to conclude that hybridization might
have played a role in speciation and that the genus might
represent a hybrid swarm.
In the cladogram based on ITS1-2 sequencing data
(Hamatani et al. 2008), a monophyletic group for x = 8
(supported with a bootstrap value of 83.3) as well as for x =
7 forming a polytomy was obtained. Two species, L. muirii
and L. pusilla both with a basic number of 7, grouped with
the x = 8 clade, but formed the base for the rest of the x = 8
species. The ITS1-2 region seemed to have more variation
in the x = 8 taxa than in the x = 7 taxa, since the clade for x
= 8 was better resolved. A similar observation was made by
Spies (2004) with the trnL-F sequences.
The good correlation between basic chromosome num108
Floriculture and Ornamental Biotechnology 6 (Special Issue 1), 98-115 ©2012 Global Science Books
Evolution and relatedness of different basic
chromosome numbers
in sequencing data) or they could be the product of mutation or putative hybridization between species in the same
geographical distribution area. Reduction in chromosome
number either by losing a chromosome or by translocation
might have contributed to speciation in these two groups.
Hancke et al. (2001) speculated that x = 7 evolved from x =
8 through a reduction in chromosome number based on the
homoeology between two chromosomes in the x = 7 and
three chromosomes in the x = 8 species studied.
Five of the nine species in the x = 7 group (L. aloides
var. aloides, L. aloides var. aurea, L. longibracteata, L.
variegata, L. viridiflora) had very similar chromosome
morphology (Hamatani et al. 2009) and seemed to be
closely related. The close relationship between L. aloides
and L. viridiflora can be confirmed from crossing data with
a success rate of between 25 and 100% depending on the
reciprocal direction (data not shown) and the production of
fertile F1 hybrids with seven bivalents in meiotic analysis
(Hancke and Liebenberg 1998).
According to (Hamatani et al. 2009) the chromosome
morphology of L. mutabilis and L. rubida were very similar,
but differed from the above group, and the authors concluded that these species probably originated from a single
ancestral species. For the purpose of this review a selection
of ITS1-2 sequences representing only those species used in
the FISH study (Hamatani et al. 2009) were obtained from
Genbank and a phylogram was constructed (Fig. 5). The
tree was drawn to scale, with branch lengths in the same
units as those of the evolutionary distances used to infer the
phylogenetic tree. The ITS phylogram yielded similar
monophyletic groupings than the ITS1-2 cladogram (Hamatani et al. 2009) and included both L. mutabilis and L.
rubida within the x = 7 clade. Both these species have a
similar branch length that was much longer than the other
species in the clade, which supported the similarity in
chromosome morphology. This relationship cannot be confirmed from crossing data (success rate of only 10%),
neither by the data presented by Spies (2004) or Hamatani
et al. (2008).
The largest number of species in Lachenalia are found
within the basic x = 7 and 8 groups. Molecular data from
ITS1-2 (Hamatani et al. 2009) and trnL-F (Spies 2004)
sequences indicated a strong relationship between these two
basic chromosome number groups and that these groups
might have evolved from a common ancestor. Cross-ability
data confirmed a relationship between these two basic
chromosome number groups with higher success rates (18%
for x = 8 crossed with x = 7), than most of the other between
group success rates (Table 2). The existence of genome
affinity indices of 0.9 in three interspecific dibasic hybrids
(Hancke et al. 2001), as discussed above, also confirmed
this relationship.
Karyomorphological data presented by Hamatani et al.
(2009) using FISH and DAPI staining to determine the
chromosomal evolution of the x = 7 and x = 8 groups confirmed the results found from both the phylogeny and the
cross-ability. The results of this study between a group of x
= 7 (consisting of L. muirii, L. aloides var. aloides, L.
aloides var. aurea, L. longibracteata, L. variegata, L. viridiflora, L. mutabilis, L. rubida, and L. pusilla) and x = 8 (consisting of L. carnosa, L. liliflora, L. namaquensis, L. splendida and L. unicolor) led to the conclusion, that there was
little morphological chromosome variation within the x = 8
group and that this group was derived from an ancestral
species followed by ongoing speciation.
The x = 7 group showed much more variation, with four
karyotype patterns indicating several morphological alterations of chromosomes within this group. This was in contrast with the ITS1-2 region data that seemed to have more
variation in the x = 8 taxa than in the x = 7 taxa, since the
clade for x = 8 was better resolved than the polytomic x = 7
clade (Hamatani et al. 2009).
Hamatani et al. (2008, 2009) suggested several theories
for the evolution of the x = 7 and 8 groups. Both groups
might have evolved from a common ancestor (as indicated
L. viridiflora
L. variegata
L. longibracteata
L. longibracteata
L. variegata
㪌㪐
L. aloides var. aurea
L. viridiflora
㫏㩷㪔㩷㪎
L. muirrii
L. aloides var. aloides
L. mutabilis
L. aloides var. aurea
L. rubida
㪏㪎 L. rubida
L. pusilla
L. liliflora
㪉㪈
L. carnosa
L. splendida
㫏㩷㪔㩷㪏
㪍㪊
L. unicolor
L. namaquensis
L. namaquensis
㪇㪅㪇㪇㪇㪌
Fig. 5 Evolutionary relationships of 17 taxa based on the ITS1-2 region. The phylogram was constructed using the Maximum Likelihood option of
MEGA 5 (Tamura et al. 2011) to compare the evolutionary development of the x = 7 and 8 groups.
109
Review of the genus Lachenalia. Kleynhans et al.
L. mutabilis
L. reflexa
㪈㪏
L. viridiflora
L. variegata
L. muirii
㪊㪌
L. aloides var. vanzyliae
㫏㩷㪔㩷㪎
L. bulbifera
㪍㪋
L. bulbifera
㪍㪈
L. carnosa (x = 8)
㪋㪉
㪊㪊
㪍㪌
L. rubida
L. variegata
L. pusilla (x = 7)
㪌㪈
L. liliflora
L. rosea
㪊㪍
L. purpureo-caerulea
㪋㪉
L. namaquensis
L. splendida
㪋㪈
㫏㩷㪔㩷㪏
L. unicolor
L. contaminata
L. pallida
㪉㪈
㪉㪇
L. pustulata
L. doleritica (x = 9)
L. latimerae (x = 9)
L. alba (x = 10)
L. comptonii (x = 10)
L. duncanii (x = 9)
L. obscura (x = 9)
L. convallarioides (x = 10)
L. neilii (x = 9)
L. minima (x = 9)
㪏㪍
L. corymbrosa
L. paucifolia
L. maughanii
㪍㪍
㪧㫉㪼㫍㫀㫆㫌㫊㫃㫐㩷㪧㫆㫃㫐㫏㪼㫅㪸
L. ensifolia
㪊㪋
㪊㪏
L. odorata
㪊㪏
L. zeyheri
L. juncifolia
㪍㪈
L. unifolia
㪐㪎
㫏㩷㪔㩷㪈㪈
L. anguinea (2n = 30+2B)
㪎㪉
㪈㪐
L. hirta
L. isopetala
L. nervosa (2n = 16/24)
㪋㪋
L. staynerii (2n = 24)
㪍㪊
L. mediana var. rogersii (x = 13)
㪍㪋
㫏㩷㪔㩷㪈㪊
L. mediana var. mediana (x = 9)
㪐㪐
㪍㪋
L. mediana var rogersii (x = 13)
Fig. 6 Evolutionary relationships of 43 taxa based on the trnL-F region (Spies 2004), inferred using the Maximum Likelihood option of MEGA 5
(Tamura et al. 2011).
The remaining two species in the x = 7 group that were
investigated (Hamatani et al. 2009), L. muirii and L. pusilla,
shared chromosomal characteristics with species in both the
x = 7 and 8 groups. The relationship to both x = 7 and 8 of L.
110
muirii and L. pusilla was confirmed by Hamatani et al.
(2008). Hamatani et al. (2009) suggested that L. pusilla
might be intermediate between the x = 7 and x = 8 group.
None of the crosses made with L. pusilla as either parent
Floriculture and Ornamental Biotechnology 6 (Special Issue 1), 98-115 ©2012 Global Science Books
L. unicolor
L. contaminata
L. unicolor
Massonia depressa
L. liliflora
L. unicolor
L. unicolor
L. pallida
L. bachmanii
Ornithogalum umbellatum
Massonia pustulata
L. namaquensis
L. hirta
2
1
L. hirta
L. muirii
L. latimerae
L. mutabilis
*L. latifolia
L. pusilla
L. pusilla
L. zeyheri
L. rosea
L. bulbifera
L. arbuthnotiae
L. rubida
L. reflexa
L. peersii
L. orchioides
3
L. juncifolia
L. longituba
L. unifolia
L. paucifolia
L. reflexa
L. ensifolia
L. algoensis
L. variegata
L. reflexa
Fig. 7 Network of Lachenalia species based on ITS data using NETWORK 4.6.1.0 (Fluxus Technology, 2012). The correct current citation of L.
latifolia (indicated with *) is L. nervosa. Colour codes: Red, x = 7; Yellow, x = 8; Blue, x = 11; Purple, 2n = 24/26/28; Grey, x = unknown. Node 1, L.
pustulata and L. purpureo-caerulea; Node 2, L. carnosa and L. splendida; Node 3, L. aloides var. aloides, L. aloides ‘Pearsonii’, L. aloides var. luteola,
L. aloides var. vanzyliae, L. aloides var. quadricolor, L. aloides var. aurea, L. viridiflora, L. orchioides var. orchioides and L. longibracteata.
were successful, neither with x = 7 nor with x = 8 species.
The cross-ability data available can thus not shed any light
on the position of L. pusilla.
There seem to be an evolutionary relationship between
some of the other basic chromosome number groups and
even with other genera. For better insight in the evolution of
the rest of the chromosome numbers, sequences from Spies
(2004) were selected to represent a broad spectrum of
chromosome numbers in the genus. Sequences were selected based on the cladogram produced by Spies (2004),
but all sequences forming a polytomy were excluded, and a
new cladogram (Fig. 6) was constructed.
Although many of the clades are not well supported, the
new trnL-F cladogram (Fig. 6) supports the suggestion that
the genus evolved from a common ancestor. The basic numbers x = 7 and 8 evolved from a common predecessor, even
though many of the clades are not well supported, thus confirming the data presented above. The higher basic numbers
(x = 9, 10, 11 and 13) form a poorly supported monophyletic clade (bootstrap value 57). It seems as if the higher
numbers evolved independently from the lower numbers in
at least two separate events. The basic numbers x = 9 and 10
forms a polytomy in the higher clade and seems to be the
bridge from the lower to the higher numbers or vice versa
(Fig. 6). Because none of the x = 9 or 10 taxa are well
resolved, this group might be a recent group. The low level
of variation in these two basic numbers indicates that evolution was recent and these numbers have not evolved into
two definite clades.
A median-joining network (Bandelt et al. 1999) was
constructed from the ITS data (Hamatani et al. 2008) (Fig.
7) as well as from 43 trnL-F sequences (Spies 2004) (Fig.
8). The trnL-F network suggests that x = 11 and x = 8 have
evolved independently from a common ancestor, and that x
= 9 and 10 could have evolved from any one of these two
numbers. The ITS network (Fig. 7) could not confirm or
reject this, due to the lack of x = 10 species and the inclusion of only a single x = 9 species. Both the networks
support a close relationship between the x = 7 and 8 groups.
The cross-ability success rate of 33% between basic x = 10
and basic x = 8 (Table 2) could be a confirmation of the
possible bridge between x = 7 and 8 and the higher numbers.
The ITS network also supported the relationship between L.
mutabilis and L. rubida (Fig. 5) and the trnL-F network
positioned L. pusilla in an ancestral position to x = 7 and 8
thus supporting the molecular cytogenetic data.
Dysploidy (through the fusion of acrocentric chromosomes at the centromere to form larger metacentric to submetacentric chromosomes) has been shown to be important
in the chromosomal evolution of other plant families, e.g.
the Commelinaceae (Jones 1976). If dysploidy is the mode
of speciation in Lachenalia a study on the chromosome
morphology of species with higher basic chromosome numbers compared to lower basic chromosome numbers could
assist in confirming the hypotheses. A study of L. latimerae
(x = 9 according to Hamatani et al. 2007) indicated that this
species has three large chromosomes, of which two are very
similar, with the third one having a satellite (Hamatani et al.
2007). The chromosome morphology thus, supports the
theory of dysploidy, but it must be further investigated with
chromosome banding techniques. A second hypothesis is
the possibility that L. latimerae could have resulted from a
hybridization event (Hamatani et al. 2007) between x = 7
and x = 11, resulting in a gametic number of n = 18. If this
111
Review of the genus Lachenalia. Kleynhans et al.
L. anguinea
L. hirta
L. zeyheri
L. juncifolia
5
L. unifolia
Massonia depressa
L. isopetala
L. comptonii
Ornithogalum umbellatum
L. corymbosa
L. paucifolia
L. maughanii
L. pusilla
L. ensifolia
L. odorata
L. contaminata
L. convallarioides
L. duncanii
L. variegata
L. mutabilis
L. rubida
L. latimerae
L. minima
L. doleritica
1
L. obscura
L. variegata
L. liliflora
L. mediana
L. rosea
L. mediana
L. mediana
L. muirii
L. reflexa
4
L. carnosa
L. pustulata
2
L. bulbifera
L. bulbifera
3
L. pallida
Fig. 8 Network of Lachenalia species based on trnL-F data using NETWORK 4.6.1.0 (Fluxus Technology, 2012). Colour codes: Red, x = 7; Yellow,
x = 8; Blue, x = 11; Light purple, 2n = 24/26/28; Dark purple, x = 9; Orange, x = 10; Diagonal crosses, x = 9 or 13; Grey, x = unknown. Node 1, L. neilii;
L. alba; Node 2, L. purpureo-caerulea; L. unicolor; Node 3, L. namaquensis; L. splendida; Node 4, L. viridiflora; L. aloides var. vanzyliae; Node 5,
Massonia pustulata; M. depressa; M. echinata; M. jasminiflora.
investigated to the same extend as x = 7, 8, 9 and 11. With
basic chromosome numbers of 5, 6, 7, 8, 9, 10, 11, 12, 13
and 15 recorded, it is still speculated whether basic numbers
of x = 5, 6, 10, 12, 13 and 15 exists.
There are very few reports for n or x = 5 in Lachenalia,
and usually when x = 5 has been reported for a species, it
was based only on one accession. Both L. violacea and L.
aloides are x = 7 species, with a single 2n = 15 reported,
indicating possible miss counts in these species. Lachenalia
mutabilis has chromosome counts of x = 5, 6 and 7. This is
the only species where numerous counts have been recorded for all three these numbers. This species is morphologically distinct and wrong identification could not attribute to the differences in counts. All reports for x = 5 for L.
mutabilis are from the same geographical distribution area
(Clanwilliam in the Western Cape Province), but there are
also reports of x = 7 from Clanwilliam. Other species from
the Clanwilliam district include x = 7 (L. elegans var.
sauveolens, L. thomasiae and L. violaceae); x = 8 (L. unicolor); x = 10 (L. marginata and L. undulata) and x = 11 (L.
hirta and L. unifolia). It was suggested that the three basic
numbers for L. mutabilis form an aneuploidy series (Spies
et al. 2000), but there is no proof of what attributed to the
chromosome diversity in this species. Based on molecular
systematics, L. mutabilis specimens always group with
other x = 7 species, regardless of their chromosome number
(Spies 2004; Hamatani et al. 2008); are karyotypically similar to L. rubida (x = 7) and has the highest number of x = 7
counts recorded, thus supporting the theory of an aneuploid
series in the species.
Johnson and Brandham (1997) studied the karyotypes
of x = 7-13 and 15, and reported that all the species studied
formed structural diploids and thus concluded that 2n = 20
rather represents a diploid based on x = 10 than a tetraploid
based on x = 5. They did state that 2n = 30 (x = 15) could be
an allotetraploid derived from taxa with x = 7 and 8, following hybridization and doubling of the chromosome
number. Considering this theory, it would be expected that x
= 10 taxa have a phylogenetic grouping either with x = 7 or
x = 8 taxa, but this have not been observed in the trnL-F
cladogram (Spies 2004). The fact that the cross-ability
theory is correct for other x = 9 species, one would expect at
least some of the x = 9 species to group with either x = 7 or
x = 11 in the chloroplast cladogram. All the x = 9 species
fall between the x = 7/8 groups and the higher numbers, but
because the trnL-F cladogram (Fig. 6) is not supported with
high bootstrap values, neither the dysploid theory nor the
hybridization theory could be proven. The trnL-F medianjoining network (Fig. 8) is inconclusive in this matter, since
the evolutionary direction for x = 9 can be from either x =
11 or x = 7/8 or both (thus hybridization).
The group x = 11 is very well supported with a bootstrap value of 94 in the trnL-F cladogram (Fig. 6), suggesting a strong relationship within this group. The close
relationship within this group is also supported by the
morphological cladogram constructed by Duncan (2005),
even though these species do not form a monophyletic
group. The evolution of x = 11 is not clear, but from the
cladograms obtained in the different studies i.e. morphological (Duncan et al. 2005), ITS (Hamatani et al., 2008) and
trnL-F (Spies 2004), x = 11 (and x = 13) is basal to the
lower numbers and it seems that species with x = 11/13 is
the intermediate between the outgroup species (which have
higher numbers) and the lower numbers in the genus. The
network drawn from the ITS sequences provides evidence
of the link between the higher basic numbers in Lachenalia
and outgroup species used in this study. The outgroup for
the ITS network (Fig. 7) is Massonia and Ornithogalum
umbellatum. The latter species has a high degree of cytogenetical variation (Czapik 1968) with numbers of 2n = 1830 and B-chromosomes reported. Hamatani et al. (2008)
obtained the ITS sequences for L. hirta (x = 11) by cloning
the maternal and paternal genomes. One genome was
cloned in some specimens and seem to have evolved from
Massonia, while the other genome have evolved from Ornithogalum this may be the reason why different specimens
form two different nodes in the network.
Existence of basic chromosome numbers
The evolution and even existence of certain chromosome
numbers (such as x = 5, 6, 12, 13 and 15) have not been
112
Floriculture and Ornamental Biotechnology 6 (Special Issue 1), 98-115 ©2012 Global Science Books
between x = 10 and 8 is relatively high could be an indication of the validity of this theory. The existence of the
basic number x = 10, however, seem to be a reality, proven
by the fact that some species has chromosome counts of 2n
= 20, 40 (L. alba) and 2n = 30, 40 (L. isopetala – not
grouped in this study) indicating the existence of polyploids.
After all the evidence, it is still not clear whether x = 5 exist
in any other species than L. mutabilis.
Reports for six species with either x = 6 or 2n = 24
were mostly based on only one accession and differed from
the majority number of counts for these species. Lachenalia
nervosa has counts of n = 8 and 2n = 24, indicating that this
species has a basic number of x = 8 and have a triploid
somatic number. Lachenalia stayneri is also 2n = 24, and
the lack of meiotic studies in this species may lead to the
conclusion that this species represents a tetraploid based on
x = 6 or also a triploid with x = 8. Therefore x = 6 should
also be considered as a basic number. Based on trnL-F
sequences, both these species indicate close relations with L.
mediana (x = 9 and 13) and do not group with x = 8 (Spies
2004). Therefore, species with 2n = 24 cannot be considered
as “typical” x = 8 species, and might even be considered as
being miss counts based on x = 13. None of the 2n = 24
species has its own monophyletic grouping and it seems as
if x = 6 does not exist except maybe in L. mutabilis.
Somatic counts of 2n = 28 and 56 have been reported
by several authors (Moffett 1936; de Wet 1957; Crosby
1986; Hancke and Liebenberg 1990; Johnson and Brandham 1997; Hamatani et al. 1998; Kleynhans and Spies
1999; Spies et al. 2002; Hamatani et al. 2007; Spies et al.
2008, 2009), but it has not been proven whether the basic
chromosome number of x = 14 exists. Somatic numbers of
2n = 28 as sole chromosome number have been reported for
L. cernua and L. longituba. Both these species were included in the basic group x = 7 for the purpose of this
review, but additional accessions of these species, as well as
meiosis and cytomorphological data will have to be studied
to determine the actual basic chromosome number.
Existence of hybrid species
The question of natural hybridization in the genus has been
raised several times. Both the morphological and trnL-F
cladograms had monophyletic groups consisting of a mixture of chromosome numbers x = 6, 7, 8, 9, 10 and 13 and
no consistent patterns regarding similar groupings. Spies
(2004) concluded that hybridization might have a role in
speciation, but it was not proven.
Some species (L. pusilla, L. rosea and L. carnosa) do
not follow the rule of grouping into monophyletic groups
with similar chromosome numbers (Fig. 6). Considering the
positions of these species in the networks drawn (Figs. 8, 9)
the first two species is intermediate to the x = 7 and x = 8
groups in both networks. The position of L. carnosa (x = 8)
fluctuate between x = 7 (Fig. 8) and x = 8 (Fig. 7). Within
the trnL-F cladogram, L. carnosa, L. rubida and L. bulbifera is a sister clade with the rest of the x = 7 species.
Lachenalia rubida is intermediate to x = 7 and 8 in both
networks. To conclude, based on karyotypic and molecular
data, some species are intermediate between x = 7 and 8,
and can either be considered as predecessor species or as
hybrid species.
Lachenalia carnosa (x = 8) is an example of a possible
hybrid species, grouping with either x = 7 or 8, depending
on the type of sequencing data (nuclear or cytoplasmic).
Spies (2004) reported what seemed to be B-chromosomes
in the meiotic divisions if L. carnosa, which may have been
unidentified univalents, also observed in cultivated Lachenalia hybrids (Hancke and Liebenberg 1998). Cross-ability
data, however, strongly links L. carnosa with other members of the x = 8 group, successfully crossing with at least
five different x = 8 species (data not shown), producing
regular meiosis with 8 bivalents (Du Preez et al. 2002) as
well as fertile hybrids. Natural hybridization may be present
in the genus Lachenalia but this should be investigated
113
further.
CONCLUSION
This review accentuates the complex nature of the genus
Lachenalia. Besides the extensive morphological variation
that complicates the taxonomy of the genus, the genus is
also exceptionally diverse in chromosome numbers. Lachenalia has different basic chromosome numbers (x = 5, 6, 7,
8, 9, 10, 11, 12, 13 and 15 reported in literature), contains
polyploidy (ranging from triploids to octoploids), and
includes B-chromosomes. Chromosome counts for the 89
species reported in literature varied from 2n = 10 to 56 and
from n = 5 to 28. Polyploidy was reported in 19 taxa (23%),
and is most common in the x = 7 group.
The low cross-ability (only 18% successful interspecies crosses) reiterates this variation and stresses the
importance of investigating the variation in order to develop
breeding strategies to overcome the existing crossing
barriers. Morphological and molecular phylogenetic studies
confirm the complexity of the genus, but also assisted in
drawing some conclusions on the relationship between species within the genus and the possible evolutionary history
of the genus.
Phylogenetic studies has assisted in finding the phylogenetic position of Lachenalia in relation to other genera
(Pfosser and Speta 1999; Pfosser et al. 2003; Manning et al.
2004) and placed the genus within the Asparagaceae family
(APG III group 2009). Morphological (Duncan et al. 2005)
and phylogenetic studies within the genus (Spies 2004;
Hamatani et al. 2008) supported the inclusion of Polyxena
in Lachenalia, and this inclusion increased the number of
recognised Lachenalia species to 126.
Molecular studies on the trnL-F as well as ITS regions
revealed monophyletic groupings of species containing the
same basic chromosome numbers. This indicated a strong
correlation between the phylogeny and basic chromosome
numbers in the genus, although there were some exceptions
in the larger trnL-F data set (Spies 2004). The good correlation between basic chromosome numbers and phylogenetic
groupings could in the future assist to confirm basic numbers for species. The improved cross-ability when crosses
were made between individual species within the same
phylogenetic groupings confirms the phylogeny. Phylogenetic groupings, thus has to be taken into account when
crossing combinations are planned to achieve better crossing success rates in the breeding programme.
When comparing the different studies, Lachenalia
might have evolved from a common ancestor and the two
largest basic chromosome number groups, x = 7 and 8 have
evolved from a common predecessor. The studies also indicated a close relationship between these two basic numbers,
which is supported by higher success rates in cross-ability
between these two groups. It seems as if the higher basic
numbers (x = 9, 10, 11 and 13) evolved independently from
the lower numbers and that basic numbers x = 9 and 10
could be the bridge from the lower to the higher numbers or
vice versa (Fig. 6), but evidence of this is not conclusive
(Figs. 7, 8).
Dysploidy and hybridization might be the modes of
speciation in some Lachenalia species but this could not be
proven with molecular data and further studies are required
to draw conclusions. The existence of some of the basic
chromosome numbers reported (such as x = 5, 6, 10, 12 and
15) can been disputed. Only a few species can be linked to x
= 5 and 6 and it is possible that these two basic numbers
only exist as part of an aneuploid series in the species L.
mutabilis. Further studies on species from these disputed
basic chromosome numbers is needed to resolve the
existence of all the reported numbers.
This review indicates that different genetic studies on
Lachenalia reveal similar results and stresses the importance of assessing the variation within complex genera to
aid in decisions around breeding programme strategies. It is
clear that inter-species crosses within phylogentic groups in
Review of the genus Lachenalia. Kleynhans et al.
the genus can improve the success rate of crossing combinations, but there are still many questions that remain unanswered. Further multidisciplinary studies are needed in
the genus Lachenalia to solve the evolutionary history of
this complex genus, to answer questions around species
placement and the existence of basic chromosome number
groups and to overcome crossing barriers.
18S rDNA probes and DAPI staining. Chromosome Botany 4, 57-63
Hamatani S, Tagashira N, Kondo K (2010) Molecular cytogenetic analysis in
seven species of Lachenalia (Liliaceae) with the chromosome numbers of
2n=18, 22, 23, 26 and 28 by DAPI staining and FISH using 5S rDNA and
18S rDNA probes. Chromosome Botany 4, 57-63
Hancke FL (1991) ‘n Sitotaksonomiese ondersoek van sewe Lachenalia spesies vir gebruik in ‘n blomteeltprogram. MSc thesis, University of Pretoria,
74 pp
Hancke FL, Liebenberg H (1990) B-chromosomes in some Lachenalia species and hybrids. South African Journal of Botany 56, 659-664
Hancke FL, Liebenberg H (1998) Meiotic studies of interspecific Lachenalia
hybrids and their parents. South African Journal of Botany 64, 250-255
Hancke FL, Liebenberg H, Janse van Rensburg WJ (2001) Chromosome
associations of three interspecific, dibasic Lachenalia hybrids. South African
Journal of Botany 67, 193-198
Johnson MAT, Brandham PE (1997) New chromosome numbers in petaloid
monocotyledons and in other miscellaneous angiosperms. Kew Bulletin 52,
121-138
Jones K (1976) Multiple Robertsonian fusions in the evolution of a plant genus.
In: Jones K, Brandham PE (Eds) Current Chromosome Research, Elsevier
North-Holland, Amsterdam, pp 220-221
Kleynhans R (2006) Lachenalia, spp. In: Anderson NO (Ed) Flower Breeding
and Genetics: Issues, Challenges, and Opportunities for the 21st Century,
Springer, Berlin, pp 491-516
Kleynhans R (2009) Back to basics for new crop development. Acta Horticulturae 836, 185-191
Kleynhans R (2011) Potential new lines in the Hyacinthaceae. Acta Horticulturae 886, 139-145
Kleynhans R, Hancke FL (2002) Problems and breeding strategies for the
development of new Lachenalia cultivars. Acta Horticulturae 570, 233-240
Kleynhans R, Spies JJ (1999) Chromosome number and morphological variation in Lachenalia bulbifera. South African Journal of Botany 64, 357-360
Kleynhans R, Spies JJ (2000) Evaluation of genetic variation in Lachenalia
bulbifera (Hyacinthaceae) using RAPDs. Euphytica 115, 141-147
Kleynhans R, Spies JJ, Spies P (2009) Cross-ability in Lachenalia. Acta Horticulturae 813, 385-392
Lubbinge J (1980) Lachenalia Breeding I. Introduction. Acta Horticulturae
109, 289-295
Manning JC, Goldblatt P, Fay MF (2004) A revised generic synopsis of Hyacinthaceae in sub-Saharan Africa, based on molecular evidence, including
new combinations and the new tribe Pseudoprosperae. Edinburgh Journal of
Botany 60, 533-568
Manning JC, Goldblatt P, Snijman D (2002) The Color Encyclopedia of Cape
Bulbs, Timber Press, Portland, Cambridge, 486 pp
Moffett AA (1936) The cytology of Lachenalia. Cytologia 7, 490-498
Mogford DJ (1978) Centromeric heterochromatin in Lachenalia tricolor (L.)
Thunb. Journal of South African Botany 44, 111-117
Moore FW (1905) Lachenalia hybrids. The Gardener’s Chronicle 37, 210-211
Müller-Doblies U, Müller-Doblies D (1997) A partial revision of the tribe
Massonieae (Hyacinthaceae). 1. Survey, including three novelties from
Namibia: A new genus, a second species in the monotypic Whiteheadia, and
a new combination in Massonia. Feddes Repertorium 108, 49-96
Müller-Doblies U, Nordenstam B, Müller-Doblies D (1987) A second species
in Lachenalia subgen. Brachyscypha (Hyacinthaceae): Lachenalia barkeriana sp. Nov. from southern Little Namaqualand. South African Journal of
Botany 53, 481-488
Nordenstam B (1982) Chromosome numbers of southern African plants 2.
Journal of South African Botany 48, 273-275
Ornduff R, Watters PJ (1978) Chromosome numbers in Lachenalia (Liliaceae). Journal of South African Botany 44, 387-390
Perry P (1985) The restructuring of the family Liliaceae. Veld and Flora Sept.,
pp 66-68
Pfosser M, Speta F (1999) Phylogenetics of Hyacinthaceae based on plastid
DNA sequences. Annals of the Missouri Botanical Garden 86, 852-875
Pfosser M, Wetschnig W, Ungar G, Prenner G (2003) Phylogenetic relationships among genera of Massonieae (Hyacinthaceae) inferred from plastid
DNA and seed morphology Journal of Plant Research 116, 115-132
Reinten EY, Coetzee JH, Van Wyk BE (2011) The potential of South African
indigenous plants for the international cut flower trade. South African Journal of Botany 77, 934-946
Riley HP (1962) Chromosome studies in some South African monocotyledons.
Canadian Journal of Genetic Cytology 4, 40-55
Salisbury RA (1866) The Genera of Plants. A Fragment Containing Part of
Liriogamae. J. v. Voorst, London
SANBI (2009) List of SA red data listed species. Available online:
http://www.sanbi.org/index.php?option=com_docman&task=documentdetails
&id=43
Spies JJ, Du Preez JL, Minnaar A, Kleynhans R (2000) Hyacinthaceae:
Chromosome studies on African plants. 13. Lachenalia mutabilis, L. pustulata and L. unicolor. Bothalia 30, 106-110
Spies JJ, Spies P, Reinecke SMC, Kleynhans R, Duncan GD, Edwards TJ
(2008) Lachenalia. In: Marhold K (Ed) IAPT/IOPB chromosome data 5.
Taxon 57, 554-555
ACKNOWLEDGEMENTS
The Agricultural Research Council and the University of the Free
State is thanked for the facilities and financial support during this
study.
REFERENCES
Angiosperm Phylogeny Group III (2009) An update of the Angiosperm
Phylogeny Group classification for the orders and families of flowering
plants: APG III. Botanical Journal of the Linnean Society 161, 105-21
Baker JG (1897) Lachenalia Jacq. In: Thistleton-Dyer WT (Ed) Flora Capensis 6, Reeve and Co., London, pp 421-436
Bandelt H-J, Forster P, Röhl A (1999) Median-joining networks for inferring
intraspecific phylogenies. Molecular Biology and Evolution 16, 37-48
Barker WF (1983) A list of the Lachenalia species included in Rudolf Schlechter’s collections made in 1891-1898 on his collection trips in southern Africa,
with identifications added. Journal of South African Botany 49, 45-55
Crosby TS (1978) Hybridisation in the genus Lachenalia. Veld and Flora Sept.,
87-90
Crosby TS (1986) The genus Lachenalia. The Plantsman 8, 129-166
Czapik R (1968) Chromosome numbers of Ornithogalum umbellatum L. from
three localities in England. Watsonia 6, 345-349
De Wet JMJ (1957) Chromosome numbers in the Scilleae. Cytologia 22, 145159
Dold AP, Phillipson PB (1998) A revision of Lachenalia (Hyacinthaceae) in the
Eastern Cape, South Africa. Bothalia 28, 141-149
Du Plessis NM, Duncan GD (1989) Bulbous Plants of Southern Africa, Tafelberg Publishers, Cape Town, pp 78-82
Du Preez JL, Spies JJ, Kleynhans R (2002) A preliminary study of interspecific hybrids in Lachenalia (Hyacinthaceae). Acta Horticulturae 570, 319-326
Duncan GD (1988) The Lachenalia handbook. Annals of Kirstenbosch Botanic
Gardens 17, 1-71
Duncan GD (1992) The genus Lachenalia: Its distribution, conservation status
and taxonomy. Acta Horticulturae 325, 843-845
Duncan GD (1996) Four new species and one new subspecies of Lachenalia
(Hyacinthaceae) from arid areas of South Africa. Bothalia 26, 1-9
Duncan GD (1998) Notes on the genus Lachenalia. Herbertia 53, 40-48
Duncan GD (2001) Lachenalia elegans var. flava (Hyacinthaceae). Curtis’s
Botanical Magazine 18, 18-22
Duncan GD (2002) Lachenalia. In: Manning J, Goldblatt P, Snijman D (Eds)
The Color Encyclopedia of Cape Bulbs, Timber Press, Cambridge, UK, pp
251-264
Duncan GD (2005) Character variation and a cladistic analysis of the genus
Lachenalia Jacq.f. ex Murray (Hyacinthaceae: Massonieae). MSc thesis, University of KwaZulu Natal, 338 pp
Duncan GD, Edwards TJ, Mitchel A (2005) Character variation and a cladistic analysis of the genus Lachenalia Jacq. f. ex Murray (Hyacinthaceae). Acta
Horticulturae 673, 113-120
Duncan GD, Edwards TJ (2006) Three new species of Lachenalia (Hyacinthaceae: Massonieae) from Western and Northern Cape, South Africa. Bothalia 36, 147-155
Duncan GD, Edwards TJ (2007) A new pyrophytic Lachenalia species (Massonieae) from Western Cape, South Africa. Bothalia 37, 31-34
Fluxus Technology Ltd. (2012) NETWORK 4.6.1.0. Shareware software
Available online: fluxus-engineering.com/sharenet.htm
Gouws JB (1965) Cytological studies in the genus Lachenalia Jacq. Annals of
the University College of the Western Cape 2, 1-7
Hamatani S, Hashimoto K, Kondo K (1998) A comparison of somatic
chromosome at metaphase in Lachenalia (Liliaceae). Chromosome Science 2,
21-25
Hamatani S, Ishida G, Hashimoto K, Kondo K (2004) A chromosome study
of ten species of Lachenalia (Liliaceae). Chromosome Science 8, 55-61
Hamatani S, Kondo K, Kodaira E, Ogawa H (2007) Chromosome morphology of 12 species and one variety of Lachenalia and five species of closely
related, allied genera (Liliaceae). Chromosome Botany 2, 79-86
Hamatani S, Masuda Y, Kondo K, Kodaira E, Ogawa H (2008) Molecular
phylogenetic relationships among Lachenalia, Massonia and Polyxena (Liliaceae) on the basis of the internal transcribed spacer (ITS) region. Chromosome Botany 3, 65-72
Hamatani S, Tagashira N, Ishida G, Kondo K (2009) Chromosome relationships among 13 species and one variety of Lachenalia (Liliaceae) with the
chromosome numbers of 2n=14 and 16 detected by FISH using 5S rDNA and
114
Floriculture and Ornamental Biotechnology 6 (Special Issue 1), 98-115 ©2012 Global Science Books
Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S (2011)
MEGA 5: Molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Molecular
Biology and Evolution 28, 2731-2739
Therman E (1956) Chromocentres in the mitosis of Liliaceae. Archiwum Societatis Zoologicae Botanicae Fennicae Vanamo 11, 189-193
Van Rooyen P, Spies JJ, Kleynhans R (2002) The species delimitation of
Lachenalia unifolia and L. hirta. Acta Horticulturae 570, 395-401
Spies JJ, Spies P, Reinecke SMC, Minnaar A, Du Preez JL, Kleynhans R
(2009) Lachenalia. In: Marhold K (Ed) IAPT/IOPB chromosome data. Taxon
58, 1288-1289
Spies JJ, Van Rooyen P, Kleynhans R (2002) The subgeneric delimitation of
Lachenalia (Hyacinthaceae). Acta Horticulturae 570, 225-232
Spies P (2004) Phylogenetic relationships of the genus Lachenalia with other
related liliaceous taxa. MSc, University of the Free State, 155 pp
Stebbins GL (1950) Variation and Evolution in Plants, Columbia University
Press, New York, 216 pp
115