3
S. Afr. J. Bot. 1996, 62(4): 217-219 Short Communications Notes on the leaf anatomy of Chortolirion and Haworthia (Aioaceae) G.F. Smith,* E.M.A. Steyn, 1 C. Steyn 1 and A. E. van Wyk 2 *Plant Systematics Research Subdirectorate, National Botanical Institute, Private Bag X 101, Pretoria, 0001 Republic of South Africa 1 Research Herbarium Support Services, National Botanical Institute, Private Bag X1 01, Pretoria, 0001 Republic of South Africa 2 Department of Botany, University of Pretoria, Pretoria, 0002 Republic of South Africa Received 30 October 1995; revised 26 March 1996 A comparison of chlorenchymatous tissue, as seen in transverse sections of leaves of Chortolirion angolense (Bak.) A.Berger and five representatives of the genus Haworthia Duval, suggests that the presence of a single layer of subepidermal palisade chloren- chyma cannot be used as a generic marker In the monotypic Chor- tolirion A. Berger. Keywords: Aloaceae, anatomy, Asphodelaceae, Chortolirion, Haworthia, leaf. *To whom correspondence should be addressed. The Aloaceae (often included in Asphodelaceae) is a fairly small family of succulent-leaved, petaloid monocots. About 450 spe- cies, ranging from tr ees over 20 m tall to miniatures barely 15 mm high, are distributed amongst seven genera. The family essentially has an Old World distribution, stretching southwards through sub-Saharan Africa from Saudi Arabia, and also occurs on so me of the Indian Ocean islands off the east coast of Africa, such as Socotra and Madagascar (see Smith & van Wyk 1991 for distribution maps of the genera). All the genera in the family are represented in southern Africa, treated here as the Flora of Southern Africa region. as traditionally defined, except Lomato- phyllum Willd., which is strictly a Mascarene and Madagascan genus. Biogeographically the number of Aloaceae genera decreases rapidly northwards into Africa, with only Al oe L. and Chortolirion A.Berger extending beyond the Kunene and Lim- popo Rivers. While there is no doubt about the monophyly of the Aloaccae, whether treated as a family or as one of two subfamilies in the Asphodelaceae, generic delimitation has been a sour ce of some controversy (e.g. Bayer 1972; Smith 1990, 1991 ). The generic status of particularly th e two monotypic genera, Chortolirion and Poellnitzia Uitewaal, has been debated from time to time (Smith 1991, 1994). In two contributions, Smith & van Wyk (1992, 1993) have argued that, apart from the aberrant bulbous and graminoid vegetative morphology of Chortolirion, its l eaf anat- omy, as seen in transverse section, provides additional support for recognizing it as a segregate haworthioid ge nu s. Th e availa- bility of fresh leaf material of the two superficially similar, grass- like species of Haworthia Duval, H. graminifolia G.G.Sm. and H. blackbumiae W.F.Barker, has now made possible a wider comparison of leaf anatomical characters as potential taxonomi c markers in Chortolirion. This contribution reports on the results of comparing some leaf anatomical characters of the monotypic Chortolirion ango- 217 lense (Bak.) A.Berger with those of th e two graminoid speci es of Hmvorthia which are both included in the subgenus Haworthia. For comparative purposes, a representative of H. subg. He:mngu- lares Uitewaal ex M.B.Baye r, namely H. koelmaniorum Oberm. & D.S.Hardy, a represe nt ative of H. subg. Rob11stipedtmculares Uitewaal ex M.B.Bayer, namely H. kingiana Poelln. and an arti- ficial inters ub generic hybrid, H. cymbiformis (Haw.) Duval [H. subg. Haworth ia] x H. attenuata (Haw.) Haw. [H. subg. Hexan- gulares], were also included in the survey. Material for the pr esent study was collected from plants in a wild population of C. cmgolense near Potchefstroom, North-West Province, as was described previous ly (Smith & van Wyk 1992, 1993), and fr om relocated plants in the Karoo National Botani cal Garden at Worcester (H. blackbumiae, H. graminifolia), or from the private collection of one of us (G.F.S.) (H. cymbiformis x H. attenuata), or in the Pretoria National Botanical Garden at Brum- meria, Pretoria (H. kingiana, H. koelmanior11m). All the species of Haworthia included in this survey are Red Data Book species and variously threatened (Hilton-Taylor & Smith 1994). Deposi- tion of voucher specimens in a herbarium was therefore impracti- cal. The corresponding author has had a long-standing interest in va rious aspects of the taxonomy and biology of the genus and confirmed the identity of the study material. Portions of leaves were fixed and stored in either formalin- acetic acid - alcohol (FAA) (C. angolense, H. cymbiformis x H. atten11ata) or in a 0.1 M phosphate buffered solution (pH 7.4) of 5% formaldehyde (Collins & MacNichol 1978) (H. blackburniae, H. graminifolia, H. koetmmziorum, H. kingiana). To improve the fixation of phe- nolic substances which might be present in the tissues, 0.5% caf - feine was added to the buffered solution, according to the method of Mueller & Greenwood ( 1977). Leaf portions were trimmed to appropriate sizes and conventional methods were used during the processing for embedding in glycol methacrylate (GMA) and sectioning at 2 Jlm. Sections were stained with the periodic acid/ Schiff's reaction (PAS) and counterstained with toluidine blue (Feder & O'Brien 1968). The leaf anatomy of Chortolirion has been discussed in detail by Smith & van Wyk (1992, 1993) and is not repeated here. Most importantly for this study, the leaves of this monotypic genus have a distinct subepidermal layer of palisade ce ll s (a single layer of radially elongated cells present adjacent to both the upper and lower epidermal layers). The rest of the chlorenchyma consists of a few layers of more or less isodiametric (spongy) cells (Figure 1 ). An identical pattern is present in H. graminifolia (Figure 2). In the case of H. blackburniae, the chlorenchyma consists of at least one layer of palisade cells, followed by layers of spongy parenchyma, of which the outermost layer appears pal- isade-like (Figure 3). The other taxa included in this survey com- pletely lack palisade cells in the chl orenchyrna, which uniformly consists of more or less isodiametric cells (Figures 4-6). The availability of material of th e two grass-like species of Haworthia has made possible a more rigorous comparison of the l eaf anatomy of these species to that of Chortolirion. Clearly, the validity of the emphasis previously placed on th e presence of a single, subepidermal layer of palisade chlorenchyma as a diag- nostic feature for separating Chortolirion (Smith & van Wyk 1992, 1993) needs further critical consideration (Figure 1 ). Simi- lar layers also occur in the two grass-leaved Haworthia species (Figures 2 & 3), although this may refl ect convergence. Neither the other species nor the intersubgeneric hybrid included in this survey have layers of subepidermal palisade cells (Figures 4-6). In contrast to Chortolirion and the grass-leaved species of Haworthia, the latter Haworthia taxa all have the typi cal, com- pact, rosulate l ea f arrangement reminiscent of a miniature aloe. It is therefore likely that the presence of palisade chlorenchyma in the graminoid haworthias and Chortolirion is rather related to

Notes on the leaf anatomy of Chortolirion and Haworthia ...Haworthia, the latter Haworthia taxa all have the typical, com pact, rosulate leaf arrangement reminiscent of a miniature

  • Upload
    others

  • View
    7

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Notes on the leaf anatomy of Chortolirion and Haworthia ...Haworthia, the latter Haworthia taxa all have the typical, com pact, rosulate leaf arrangement reminiscent of a miniature

S. Afr. J. Bot. 1996, 62(4): 217-219

Short Communications

Notes on the leaf anatomy of Chortolirion and Haworthia (Aioaceae)

G.F. Smith,* E.M.A. Steyn, 1 C. Steyn1 and A. E. van Wyk2

*Plant Systematics Research Subdirectorate, National Botanical Institute, Private Bag X 101, Pretoria, 0001 Republic of South Africa 1Research Herbarium Support Services, National Botanical Institute, Private Bag X1 01, Pretoria, 0001 Republic of South Africa 2Department of Botany, University of Pretoria, Pretoria, 0002 Republic of South Africa

Received 30 October 1995; revised 26 March 1996

A comparison of chlorenchymatous tissue, as seen in transverse sections of leaves of Chortolirion angolense (Bak.) A.Berger and five representatives of the genus Haworthia Duval, suggests that the presence of a single layer of subepidermal palisade chloren­chyma cannot be used as a generic marker In the monotypic Chor­tolirion A. Berger.

Keywords: Aloaceae, anatomy, Asphodelaceae, Chortolirion, Haworthia, leaf.

*To whom correspondence should be addressed.

The Aloaceae (often included in Asphodelaceae) is a fairly small family of succulent-leaved, petaloid monocots. About 450 spe­cies, ranging from trees over 20 m tall to miniatures barely 15 mm high, are distributed amongst seven genera. The family essentially has an Old World distribution, stretching southwards through sub-Saharan Africa from Saudi Arabia, and also occurs on some of the Indian Ocean islands off the east coast of Africa, such as Socotra and Madagascar (see Smith & van Wyk 1991 for distribution maps of the genera). All the genera in the family are represented in southern Africa, treated here as the Flora of Southern Africa region. as traditionally defined, except Lomato­phyllum Willd., which is strictly a Mascarene and Madagascan genus. Biogeographically the number of Aloaceae genera decreases rapidly northwards into Africa, with only Aloe L. and Chortolirion A.Berger extending beyond the Kunene and Lim­popo Rivers.

While there is no doubt about the monophyly of the Aloaccae, whether treated as a family or as one of two subfamilies in the Asphodelaceae, generic delimitation has been a source of some controversy (e.g. Bayer 1972; Smith 1990, 1991 ). The generic status of particularly the two monotypic genera, Chortolirion and Poellnitzia Uitewaal, has been debated from time to time (Smith 1991, 1994). In two contributions, Smith & van Wyk (1992, 1993) have argued that, apart from the aberrant bulbous and graminoid vegetative morphology of Chortolirion, its leaf anat­omy, as seen in transverse section, provides additional support for recognizing it as a segregate haworthioid genus. The availa­bility of fresh leaf material of the two superficially similar, grass­like species of Haworthia Duval, H. graminifolia G.G.Sm. and H. blackbumiae W.F.Barker, has now made possible a wider comparison of leaf anatomical characters as potential taxonomic markers in Chortolirion.

This contribution reports on the results of comparing some leaf anatomical characters of the monotypic Chortolirion ango-

217

lense (Bak.) A.Berger with those of the two graminoid species of Hmvorthia which are both included in the subgenus Haworthia. For comparative purposes, a representative of H. subg. He:mngu­lares Uitewaal ex M.B.Bayer, namely H. koelmaniorum Oberm. & D.S.Hardy, a representative of H. subg. Rob11stipedtmculares Uitewaal ex M.B.Bayer, namely H. kingiana Poelln. and an arti­ficial intersubgeneric hybrid, H. cymbiformis (Haw.) Duval [H. subg. Haworthia] x H. attenuata (Haw.) Haw. [H. subg. Hexan­gulares], were also included in the survey.

Material for the present study was collected from plants in a wild population of C. cmgolense near Potchefstroom, North-West Province, as was described previously (Smith & van Wyk 1992, 1993), and from relocated plants in the Karoo National Botanical Garden at Worcester (H. blackbumiae, H. graminifolia), or from the private collection of one of us (G.F.S.) (H. cymbiformis x H. attenuata), or in the Pretoria National Botanical Garden at Brum­meria, Pretoria (H. kingiana, H. koelmanior11m). All the species of Haworthia included in this survey are Red Data Book species and variously threatened (Hilton-Taylor & Smith 1994). Deposi­tion of voucher specimens in a herbarium was therefore impracti­cal. The corresponding author has had a long-standing interest in various aspects of the taxonomy and biology of the genus and confirmed the identity of the study material. Portions of leaves were fixed and stored in either formalin- acetic acid - alcohol (FAA) (C. angolense, H. cymbiformis x H. atten11ata) or in a 0.1 M phosphate buffered solution (pH 7.4) of 5% formaldehyde (Collins & MacNichol 1978) (H. blackburniae, H. graminifolia, H. koetmmziorum, H. kingiana). To improve the fixation of phe­nolic substances which might be present in the tissues, 0.5% caf­feine was added to the buffered solution, according to the method of Mueller & Greenwood ( 1977). Leaf portions were trimmed to appropriate sizes and conventional methods were used during the processing for embedding in glycol methacrylate (GMA) and sectioning at 2 Jlm. Sections were stained with the periodic acid/ Schiff's reaction (PAS) and counterstained with toluidine blue (Feder & O'Brien 1968).

The leaf anatomy of Chortolirion has been discussed in detail by Smith & van Wyk (1992, 1993) and is not repeated here. Most importantly for this study, the leaves of this monotypic genus have a distinct subepidermal layer of palisade cells (a single layer of radially elongated cells present adjacent to both the upper and lower epidermal layers). The rest of the chlorenchyma consists of a few layers of more or less isodiametric (spongy) cells (Figure 1 ). An identical pattern is present in H. graminifolia (Figure 2). In the case of H. blackburniae, the chlorenchyma consists of at least one layer of palisade cells, followed by layers of spongy parenchyma, of which the outermost layer appears pal­isade-like (Figure 3). The other taxa included in this survey com­pletely lack palisade cells in the chlorenchyrna, which uniformly consists of more or less isodiametric cells (Figures 4-6).

The availability of material of the two grass-like species of Haworthia has made possible a more rigorous comparison of the leaf anatomy of these species to that of Chortolirion. Clearly, the validity of the emphasis previously placed on the presence of a single, subepidermal layer of palisade chlorenchyma as a diag­nostic feature for separating Chortolirion (Smith & van Wyk 1992, 1993) needs further critical consideration (Figure 1 ). Simi­lar layers also occur in the two grass-leaved Haworthia species (Figures 2 & 3), although this may reflect convergence. Neither the other species nor the intersubgeneric hybrid included in this survey have layers of subepidermal palisade cells (Figures 4-6). In contrast to Chortolirion and the grass-leaved species of Haworthia, the latter Haworthia taxa all have the typical, com­pact, rosulate leaf arrangement reminiscent of a miniature aloe. It is therefore likely that the presence of palisade chlorenchyma in the graminoid haworthias and Chortolirion is rather related to

Page 2: Notes on the leaf anatomy of Chortolirion and Haworthia ...Haworthia, the latter Haworthia taxa all have the typical, com pact, rosulate leaf arrangement reminiscent of a miniature

2 18

function and growth habit, which curtails its usefulness as a dis­tinguishing character. There is also a distinct possibility that the palisade layer in the leaves of C. angofense, H. gmminijolia and

---I

/ -{

S. Afr. J. Bot. 1996, 62( 4)

H. blackburniae could have been independently acquired, spe­cific autapomorphies through parallel development - a view that we tend to favour. Regardless of the apparent lack of unequivocal

Figures 1- 6 Adaxial epidermis and adjacent chlorenchyma as seen in transverse sections of leaves of: 1. Chortolirion angolense; 2. Hawor­thia graminifolia; 3. H. blackbumiae; 4. H. cymbifonnis x H. attenuata; 5. H. koelmaniorum and 6. H. kingiana. The first three species have comparatively thin, grass-like leaves, whereas the last three are compact, miniature, rosulate leaf succulents. Scale marker = 50 !-liD.

Page 3: Notes on the leaf anatomy of Chortolirion and Haworthia ...Haworthia, the latter Haworthia taxa all have the typical, com pact, rosulate leaf arrangement reminiscent of a miniature

S. Afr. J. Bot. 1996, 62(4): 219- 22 1

leaf anatomical evidence for regarding Chortolirion as a mono­type, it is proposed that the genus should s till be maintained sep­arate from Haworthia. This stems primarily from its bulbous vegetative morphology and acuminate capsules, both features being absent in Haworthia. Futhermore, it is the only hawor­thioid species of which the leaves are deciduous and die back to ground level after tires or frost.

It is, however, clear that the outer periclinal epidermal cell walls of Chortolirion are strongly PAS-positive, whereas, in H. blackbumiae and H. graminifolia, the corresponding walls are only slightly so (Figures 1-3), indicating a general lack of com­plex polysaccharides in the graminoid haworthias. A difference in tixativcs could not have been a causal factor; the epidermal layers of H. kingiana and H. koelmaniorum showed PAS-posi­tivity after treatment with the same fixative as the graminoid haworthias (compare Figures 5 & 6 with Figures 2 & 3). Although this feature could be an additional diagnostic character for separating Chortolirion from grass-leaved species of Hawor­thia, a more comprehensive and detailed study is required before inferences can be made of its taxonomic value.

Acknowledgements

Mr Ian Oliver, curator of the Karoo National Botanical Garden, Worcester, is thanked for providing material of H. blackburniae and H. graminifolia.

References BAYER, M.B. 1972. Reinstatement of the genus Astmloba and Poe/1-

nitzia Uitw. (Li liaceae-Aloineae). Nat. Caa Sw:c. J. 27: 77-79. COLLINS. B.A . & MACNICHOL. E.F. (jun.) 1978. Long term fixation

for histological studies. Mic:m.f''· Acta 81: 155-158. FEDER, N. & O'BRIEN. T.P. 1968. Plant microtechnique: some princi­

ples and new methods. Am. J. Bot. 55: 123-142. HILTON-TAYLOR. C. & SMITH, G.F. 1994. The conservation status of

Alonceae in southern Africa. In: Botanical diversity in southern Africa. cd. B.J. Huntley. pp. 287- 303. Strelitzia I. National Botanical Institute, Pretona.

MUELLER. W.C. & GREENWOOD, A.D. 1977. The ultrastructure of phenolic c~lls fixed with caffeine. J. exp. Bot. 29: 757- 764.

SMITH. G.F. 1990. Nomenclatural notes on the subsection Bowieae in Aloe (Asphodelacene: Alooideae). S. Afr. J. Bot. 56: 303-308.

SMITH. G.F. 199 1. Studies on the reproductive biology and palynology of Clwrtolirimz Berger (Asphodelacene: Aloacene) in southern Africa. Taxnn40: 61 - 73.

SMITH. G. F. 1994. Taxonomic history of Poellnitzia Uitcwaal. a unispe­cific genus of A looidene (Asphodelnceae). Haselumia 2: 74- 78.

SMITH. G.F. & VAN WYK. A.E. 1992. Systematic leaf anatomy of selected genera of southern African Alooideae (Asphodelaceae). S. Afr. J. Hot. Sll: 349- 357. 543.

SMITH. G.F. & VANWYK. A .E. 1993. The generic status of Clwrtolir­inn (Aloaccae). Kew Bull. 48: 105-1 13.

SMITH. G.F. & VAN WYK, B-E. 1991. Generic relationships in the Alooidcac (Asphodelnccac). Taxon 40: 557-581.

219

In vitro propagation of Kniphofia pauci­flora Bak. for conservation purposes

B.G. McAlister and J. van Staden* Natal University Research Unit for Plant Growth and Development. Department of Botany, University of Natal Pietermaritzburg, Private Bag X01, Scottsville, 3209 Republic of South Africa

Received 23 January / 996: revised 27 April 1996

In vitro shoot formation was initiated using the apical regions of a stolon of Kniphofia pauciflora. Murashige and Skoog medium sup­

plemented with 100 mg 1·1 myo-inositol, 3% sucrose and solidified

with 0.8% agar was used. Kinetin (2 mg 1'1) and NAA (1 mg t·1)

were used to obtain shoot development. Shoots were subse­quently placed onto hormone-free medium for rooting. Plantlets were successfully hardened off in a seedling mix in the mist house for 2 weeks, whereafter they were placed into a greenhouse and watered twice weekly.

Keywords: Indigenous, Kniphofia pauciflora, micropropagation, rare, tissue culture.

*To whom corresprmden,·e should be addressed.

Kniphojia pauciflora Bak. is a pretty plant belonging to the sub­family Asphodeloideae (Dahlgren et at. 1985). The plants have short stolons, forming groups of short stems. The 20-35-cm leaves are soft and yellow green. The leaves are 2- 8 mm wide and are triangular in cross section (Codd 1968). The flower spike is 30- 50 em long and is raised above the leaves. The inflores­cence is yellow and few-flowered compared to other Kniphofia sp. (Figure l A). These plants flower from September to Novem­ber.

Kniphofia paucijlora is known only from restricted areas near Durban in Natal, where it used to grow in marshy grasslands. This species is threatened due to urbanization and is no longer found at the type locality. Specimens were found in the Marian Hill area in 1951 and 1954 but the habitat has since been drained and the plants have disappeared from this region (Codd 1968). Conservation of this plant is needed and tissue cul ture was undertaken to increase the limited numbers held in the National Botanical Garden in Pietennaritzburg. Only one reference to the in vitro culture of Kniphofia species could be found (Nayak & Sen 1990).

A single plant was obtained from the Botanical Garden in Pietermaritzburg to use as explant source. First attempts at cul­turing this plant were undertaken using leaf and peduncle explants. The material was sterilized for 5 and 10 min using 3.5% sodium hypochlorite and then placed on Murashige and Skoog (1962) medium (MS) soliditied with 0.8% Unilab agar. The pH was adjusted to 5.8 with I M NaOH prior to being auto­claved. 1-Naphthaleneacetic acid (NAA) ( 1 mg J·1) plus 6-ben­zylaminopurine (BA) (2 mg J·1) or kinetin (2 mg 1·1) were the two hormone combinations added to the MS media. The explants were kept in constant high light intensity (33 1-1mol m·2 s·2) at 25°C. No contamination occurred. Fifteen percent of the leaf explants produced wound callus (Figure lB) but this soon browned and died. The peduncle explants also turned brown and died.

One stolon of the plant was then used as explant source. The stolon was dipped in 70% ethanol for I min fo llowed by 5 min in 0.2% Benlate and lastly, in fu ll strength Jik (3.5% NaOCI) for 10 min. The stolon was then rinsed three times in sterile water and cut up and placed onto MS medium supplemented with the same