View/Open - ResearchSpace - University of KwaZulu-Natal

Propagation of Romulea species 

Pierre André Swart 

Submitted in fulfillment of the academic requirements for the degree of 

Doctor of Philosophy 

in the Discipline of Botany 

Research Centre for Plant Growth and Development, 

School of Biological and Conservation Sciences, 

University of KwaZulu-Natal, 

Pietermaritzburg 

March 2012 

Supervisor: Prof. J. van Staden 

Co-supervisors: Dr. M.G. Kulkarni, Dr. M.W. Bairu and Prof. J.F. Finnie

Contents 

Abstract v 

Declaration viii 

Acknowledgements xi 

Publications from this Thesis xii 

Conference Contributions xii 

List of Figures xiii 

List of Tables xviii 

List of Abbreviations xx 

Namakwaland: A poem by my late father xxi 

1 Introduction 1 

1.1 PROPAGATION OF ROMULEA SPECIES FOR 

HORTICULTURAL AND CONSERVATION PURPOSES 1 

1.2 AIMS AND HYPOTHESES 3 

1.3 GENERAL OVERVIEW OF THESIS CONTENT 4 

2 Literature review 6 

2.1 MORPHOLOGY, DISTRIBUTION AND HABITAT 6 

2.1.1 Species specific morphology and distribution 12 

2.2 PHYLOGENY AND TAXONOMY 22 

2.3 CONSERVATION STATUS 23 

2.4 THE CLIMATE OF ROMULEA SPP. HABITATS 24 

2.5 SOIL SAMPLING AND ANALYSIS 35 

2.5.1 Physical properties of soil 36 

2.5.2 Organic matter 39 

2.5.3 Soil nutrients 39 

2.5.4 pH 40 

2.5.5 Salinity 41 

2.5.6 Cation and anion exchange capacity and surface charges 41 

2.5.7 Soils of Namaqualand 41 

2.5.8 Soils of Nieuwoudtville 42 

2.6 PROPAGATION OF ROMULEA SPECIES 43 

i

Contents 

2.7 GERMINATION PHYSIOLOGY 43 

2.7.1 Seed structure 43 

2.7.2 Seed germination 45 

2.7.3 Measuring germination 47 

2.7.4 Promotion and inhibition of germination 48 

2.7.5 Phytochromes and light quality 53 

2.7.6 Scarification 53 

2.7.7 Seed dormancy and the influence of temperature and 

stratification 54 

2.7.8 Seed longevity and viability 57 

2.7.9 After-ripening 59 

2.7.10 Embryo-excision as a tool for investigating mechanisms 

behind dormancy and testing viability 60 

2.7.11 Germination, dormancy and germination ecology in 

Iridaceae 61 

2.7.12 Embryo and seedling morphology of Iridaceae 61 

2.8 BRIEF REVIEW OF IN VITRO CULTURE 62 

2.8.1 Explant selection 64 

2.8.2 Explant preparation 65 

2.8.3 Medium composition 67 

2.8.4 Liquid culture 75 

2.8.5 Embryo-excision 75 

2.8.6 Callus culture 77 

2.8.7 Organogenesis 79 

2.8.8 Somatic embryogenesis 80 

2.8.9 Hardening 81 

2.8.10 Applications of in vitro culture 83 

2.9 CORM PHYSIOLOGY 83 

2.10 IN VITRO FLOWERING 84 

2.11 IN VITRO PROPAGATION OF GEOPHYTES 85 

2.12 IN VITRO PROPAGATION OF BULBOUS PLANTS 86 

2.13 IN VITRO PROPAGATION OF IRIDACEOUS SPECIES 86 

ii

3 Investigation into the habitat of Romulea sabulosa and Romulea 

Contents 

monadelpha: Soil sampling and analysis 92 

3.1 INTRODUCTION 92 

3.2 MATERIALS AND METHODS 93 

3.3 RESULTS 94 

4.4 DISCUSSION 97 

4.5 SUMMARY 97 

4 Germination physiology 98 



4.2.1 Viability tests 99 

4.2.2 Water content and imbibition rate 100 

4.2.3 Scanning electron microscopy 100 

4.2.4 Ex vitro germination experiments 100 

4.2.5 In vitro germination experiments 102 

4.2.6 Statistical analysis 102 


4.3.1 Viability tests 103 

4.3.2 Water content and imbibition rate 103 

4.3.3 Scanning electron microscopy 105 

4.3.4 Ex vitro germination experiments 108 

4.3.5 In vitro germination experiments 110 



5 In vitro culture initiation and multiplication 115 



5.2.1 Explants from seedlings 116 

5.2.2 Explants from embryos 117 

5.2.3 Explant comparison 120 

5.2.4 Shoot multiplication 120 

5.2.5 Statistical analysis 121 


iii

Contents 

5.3.1 Explants from seedlings 121 

5.3.2 Explants from embryos 122 

5.3.3 Explant comparison 130 

5.3.4 Shoot multiplication 132 



6 In vitro corm formation and flowering and ex vitro acclimatization 



6.2.1 Corm formation 138 

6.2.2 In vitro flowering 140 

6.2.3 Ex vitro acclimatization and corm viability 140 


6.3.1 Corm formation 142 

6.3.2 In vitro flowering 145 

6.3.3 Ex vitro acclimatization and corm viability 145 



7 Commercialization potential of Romulea species 151 

8 Literature cited 155 

iv

Abstract 

Romulea is a genus with numerous attractive and endangered species with 

horticultural potential. This genus in the Iridaceae has its centre of diversity in the 

winter-rainfall zone of South Africa. This thesis uses ecophysiological and 

biotechnological techniques to investigate the physiology behind the propagation of 

some species in this genus. 

The ecophysiological techniques of soil sampling and analysis and germination 

physiology were used to determine the natural and ex vitro growth and development 

requirements of these plants, while biotechnological techniques are used to 

determine the in vitro growth and development requirements of these plants and to 

increase the rate of multiplication and development. 

Soil sampling and analysis revealed that R. monadelpha and R. sabulosa, two of the 

most attractive species in the genus, grow in nutrient poor 1:1 mixture of clay and 

sandy loam soil with an N:P:K ratio of 1.000:0.017:0.189 with abundant calcium. 

To investigate the physical properties of the seeds, imbibition rate, moisture content 

and viability of seeds were determined. The seed coat and micropylar regions were 

examined using scanning electron microscopy. To test for suitable stimuli for 

germination, the effect of temperature and light, cold and warm stratification, acid and 

sand paper scarification, plant growth promoting substances, deficiency of nitrogen, 

phosphorous and potassium, and different light spectra (phytochromes) on 

germination were examined. An initial germination experiment showed germination 

above 65% for R. diversiformis, R. leipoldtii, R. minutiflora and R. flava seeds placed 

at 15°C; while seeds of other species placed at 15°C all had germination 

percentages lower than 30%. More extensive germination experiments revealed that 

R. diversiformis and R. rosea seed germinate best at 10°C, R. flava seed germinates 

best when cold stratified (5°C) for 21 days and R. monadelpha germinates best at 

15°C in the dark. Seeds of R. diversiformis, R. flava, R. leipoldtii, R. minutiflora, R. 

monadelpha and R. sabulosa seem to all exhibit non-deep endogenous 

morphophysiological dormancy while seeds of R. camerooniana and R. rosea appear 

to have deep endogenous morphophysiological dormancy. 

v

Abstract 

The suitability of various explant types and media supplementations for culture 

initiation was examined for various species of Romulea. Both embryos and seedling 

hypocotyls can be used for R. flava, R. leipoldtii and R. minutiflora in vitro shoot 

culture initiation. R. sabulosa shoot cultures can only be initiated by using embryos 

as explants, because of the lack of seed germination in this species. Shoot cultures 

of R. diversiformis, R. camerooniana and R. rosea could not be initiated due to the 

lack of an in vitro explant shooting response. Shoot cultures can be initiated on 

media supplemented with 2.3 to 23.2 M kinetin for all species that showed an in 

vitro response. The most suitable concentration depended on the species used. 

Some cultures appeared embryogenic, but this was shown not to be the case. A 

medium supplemented with 2.5 M mTR is most suitable for R. sabulosa shoot 

multiplication. BA caused vitrification of shoots in all the experiments in which it was 

included and is not a suitable cytokinin for the micropropagation of these species. 

The effect of various physical and chemical parameters on in vitro corm formation 

and ex vitro acclimatization and growth was examined. Low temperature significantly 

increased corm formation in R. minutiflora and R. sabulosa. A two step corm 

formation protocol involving placing corms at either 10 or 20°C for a few months and 

then transferring these cultures to 15°C should be used for R. sabulosa. When 

paclobutrazol and ABA were added to the medium on which R. minutiflora shoots 

were placed, the shoots developed corms at 25°C. This temperature totally inhibits 

corm formation when these growth retardants are not present. BA inhibited corm 

formation in R. leipoldtii. Corms can be commercialized as propagation units for 

winter-rainfall areas with minimum temperatures below 5°C during winter. 

Although an incident of in vitro flowering was observed during these experiments, 

these results could not be repeated. Although none of the corms or plantlets planted 

ex vitro in the greenhouse survived, a small viability and an ex vitro acclimatization 

experiment shows that the corms produced in vitro are viable. 

One embryo of the attractive R. sabulosa, produces 2.1 ± 0.7 SE shoots after 2 

months; subsequently placing these shoots on a medium supplemented with 2.5 µM 

mTR for a further 2 months multiplies this value by 5.5 ± 1.3 SE. Each of these 

shoots can then be induced to produce a corm after 6 months. This means that 1 

vi

Abstract 

embryo can produce about 12 corms after 10 months or about 65 corms after 12 

months (if shoots are subcultured to medium supplemented with 2.5 µM mTR for 

another 2 months). Embryo rescue can enable wider crosses within this genus. 

These results can be used for further horticultural development of species in this 

genus and their hybrids and variants. 

vii

Declarations 

I Pierre André Swart, student number 207519473, hereby declare that: 

• This thesis, Propagation of Romulea species, unless otherwise acknowledged 

to the contrary in the text, is the result of my own investigation, under the 

supervision of Professor J. van Staden and co-supervision of Doctor M.G. 

Kulkarni, Doctor M.W. Bairu and Professor J.F. Finnie, in the Research Centre 

for Plant Growth and Development, School of Biological and Conservation 

science, University of KwaZulu-Natal, Pietermaritzburg; 

• This dissertation has not been submitted for any degrees or examination at 

any other university; 

• This thesis does not contain data, figures or writing, unless specifically 

acknowledged, copied from other researchers. Where other written sources 

have been quoted, then 

1. Their words have been re-written but the general information attributed to 

them has been referenced 

2. Where their exact words have been used, then their writing has been 

placed in italics and inside quotation marks, and referenced, and; 

• Where I have reproduced a publication of which I am an author or co-author, I 

have indicated which part of the publication was contributed by me. 

• This thesis does not contain text, graphics or tables copied and pasted from 

the internet, unless specifically acknowledged, and the source being detailed 

in the thesis and in the References sections. 

Signed at on the day of 

, 2012. 


viii

Declarations 

We declare that we have acted as supervisors for this Pierre André Swart, student 

number 207519473 during this PhD study entitled Propagation of Romulea species. 

Regular consultation took place between the student and ourselves throughout the 

investigation. We advised to the best of our ability and approved the final document 

for submission to the Faculty of Science and Agriculture Higher Degrees Office for 

examination by the University appointed Examiners. 

Professor J. van Staden 

Supervisor 

Doctor M.G. Kulkarni 

Co-supervisor 

Doctor M.W. Bairu 

Co-supervisor 

Professor J.F. Finnie 

Co-supervisor 

ix

Declarations 

DETAILS OF CONTRIBUTION TO PUBLICATIONS that form part and/or include 

research presented in this thesis: 

PUBLICATION 1: 

ASCOUGH, G. D., SWART, P. A., FINNIE, J. F., and VAN STADEN, J. (2011). 

Micropropagation of Romulea minutiflora, Sisyrinchium laxum and Tritonia 

gladiolaris — Iridaceae with ornamental potential. South African Journal of 

Botany 77: 216-221. 

I supplied the data on micropropagation of Romulea minutiflora and some editorial 

help. Dr. Ascough supplied all other data and did all other editing, as he is the first 

author. Prof. Finnie and Prof. van Staden were our co-supervisor and supervisor 

respectively at the time of this project. 


SWART, P. A., KULKARNI, M.G., BAIRU, M.W., FINNIE, J. F., and VAN STADEN, 

J. (2012). Micropropagation of Romulea sabulosa. Scientia Horticulturae 135: 

151-156. 

All the data and text of this paper I generated myself, Dr. Kulkarni, Dr. Bairu and Prof. 

Finnie were my co-supervisors and Prof. van Staden was my supervisor and they 

therefore supplied some editorial help. 


SWART, P. A., KULKARNI, M.G., FINNIE, J. F., and VAN STADEN, J. (2011). 

Germination physiology of four African Romulea species. Seed Science and 

Technology 39: 354-363. 

All the data and text of this paper I generated myself, Dr. Kulkarni and Prof. Finnie 

were my co-supervisors and Prof. van Staden was my supervisor and they therefore 

supplied some editorial help. 


x

Acknowledgements 

I am grateful to: 

• My supervisor and co-supervisors who gave me the opportunity to work with 

these beautiful plants and the encouragement, support and advice needed to 

complete this study and who critically reviewed my manuscripts. 

• Mr. E. Marinus, who was kind enough to share his knowledge of R. sabulosa 

and R. monadelpha propagation with me 

• Dr. J.C. Manning, who allowed me to use and modify his pictures of Romulea 

flowers 

• To all the centre members and friends who assisted and supported me in all 

kinds of ways during my time at the RCPGD. 

• To my mother for always being there for me, no matter what the situation 

For financial assistance I would like to thank: 

• My mother 

• The National Research Foundation 

• Professor J. van Staden 

xi

Publications from this thesis 




Botany 77: 216-221. 


Micropropagation of Romulea sabulosa. Scientia Horticulturae 135: 151-156. 


Germination physiology of four African Romulea species. Seed Science and 

Technology 39: 354-363. 

Conference contributions 

SWART, P.A., FINNIE J.F., and VAN STADEN, J. (2009). Thirty-fifth Annual 

Conference of the South African Association of Botanists (SAAB). University of 

Stellenbosch, Stellenbosch. 

ix

List of Figures 

Figure 2.1: Map showing the distribution of seven of the species used in 

propagation experiments. The inset of the globe in the top right corner 

indicates the location of this map on the African continent with a rectangle. 

Modified from DE VOS (1972; 1983). 

Figure 2.2: Life cycle of Romulea sabulosa, a species endemic to the winterrainfall 

area of South Africa (Modified from ASCOUGH (2008); DE VOS 

(1972); and photographs taken by Dr. John C. Manning). 

Figure 2.3: Life cycle of Romulea monadelpha, another species endemic to 

the winter-rainfall area of South Africa (Modified from ASCOUGH (2008); DE 

VOS (1972); and photographs taken by Dr. John C. Manning). 

Figure 2.4: Life cycle of Romulea camerooniana, a species occurring in 

summer-rainfall regions of Africa (Modified from ASCOUGH (2008); DE VOS 

(1972) and photographs taken by Dr. John C. Manning). 

Figure 2.5: Royal National Park weather station (28° 57’ E, 28° 41’ S, 1392 m 

above sea level) average daily minimum and maximum monthly temperatures 

(Error bars indicate standard error of the mean of last 5 years). 


above sea level) average total monthly rain (Error bars indicate standard error 

of the mean of last 5 years). 


above sea level) average daily relative humidity (Error bars indicate standard 

error of the mean of last 3 years). 

Figure 2.8: Calvinia (19° 56’ E, 31° 29’ S, 977 m above sea level) average 

daily minimum and maximum monthly temperatures (Error bars indicate 

standard error of the mean of last 5 years). 

Figure 2.9: Calvinia weather station (19° 56’ E, 31° 29’ S, 977 m above sea 

level) average total monthly rain (Error bars indicate standard error of the 

mean of last 5 years). 

Figure 2.10: Calvinia weather station (19° 56’ E, 31° 29’ S, 977 m above sea 

level) average daily relative humidity (Error bars indicate standard error of the 


Figure 2.11: Sutherland weather station (20° 4’ E, 32° 24’ S, 1458 m above 

sea level) average daily minimum and maximum monthly temperatures (Error 

bars indicate standard error of the mean of last 5 years). 


sea level) average total monthly rain (Error bars indicate standard error of the 


7 

8 

9 

10 

27 

27 

27 

28 

28 

28 

29 

29 

xiii



sea level) average daily relative humidity (Error bars indicate standard error of 

the mean of last 5 years). 

Figure 2.14: Fraserburg weather station (31° 55’ S 21° 30’ E, 1267 m above 









Figure 2.1: Beaufort West weather station (22° 35’ E, 32° 21’ S, 899 m above 









Figure 2.20: Nieuwoudville weatherstation (19° 53’ E, 31° 21’ S, 731 m above 



Figure 2.21: Nieuwoudville weather station (19° 53’ E, 31° 21’ S, 731 m above 



Figure 2.22: Nieuwoudville weather station (19° 53’ E, 31° 21’ S, 731 m above 



Figure 2.23: Malmesbury weather station (18° 43’ E, 33° 28’ S, 108 m above 









29 

30 

30 

30 

31 

31 

31 

32 

32 

32 

33 

33 

33 

xiv


Figure 2.26: Grahamstown weather station (26° 30’ E, 33° 17’ S, 642 m above 









Figure 2.29: Diagrammatic representation of a small cluster of soil illustrating 

the complexity of organic soil. Also note the air spaces between the various 

components illustrated. Modified from descriptions of SLEEMAN & BREWER 

(1988). 

Figure 2.30: A textural triangle showing the range of variation in sand, silt, and 

clay for each soil textural class (Modified from DONAHUE et al. (1983) and 

LOVELAND & WHALLEY (1991)) 

Figure 2.31: The triphasic pattern of water uptake by germinating seeds, with 

arrow showing the time of radicle protrusion (BEWLEY & BLACK, 1994). 

Figure 3.1: The colour and structure of samples one and two. Horizontal bar = 

20 mm. 

Figure 4.1: Water content (value at day zero) and imbibition rates of seeds of 

eight species of Romulea. Error bars indicate standard error of the mean. 

Figure 4.2: Scanning electron microscopic images of seeds arranged from the 

smallest to the largest for size comparison. Romulea leipoldtii (A); R. flava (B); 

R. minutiflora (C); R. sabulosa (D); R. camerooniana (E); R. rosea (F); R. 

diversiformis (G); R. monadelpha (H). Horizontal bar = 1 mm. 

Figure 4.3: Scanning electron micrographs of the seed surfaces of Romulea 

camerooniana (A); R. diversiformis (B); R. flava (C); R. leipoldtii (D); Romulea 

minutiflora (E); R. monadelpha (F); R. rosea (G) and R. sabulosa (H). 

Horizontal bar = 10 µm (the same magnification was used for all species). 

Figure 4.4: Scanning electron micrographs of the micropylar regions of seeds 

of Romulea camerooniana (A); R. diversiformis (B); R. flava (C); R. leipoldtii 

(D); R. minutiflora (E); R. monadelpha (F); R. rosea (G) and R. sabulosa (H). 

Horizontal bar = 20 µm. 

Figure 4.5: Effect of nutrients without N, P or K, plant growth promoting 

substances and smoke constituents on seed germination of Romulea rosea 

under 16 h photoperiod at 20 ± 0.5°C. A number above the standard error bar 

represents mean germination time and an asterisk denotes that the treatment 

34 

34 

34 

36 

38 

46 

95 

104 

105 

105 

106 

xv

was significantly different from the control (water) according to LSD test at the 

5% level. 


Figure 4.6: In vitro seed germination of different Romulea species at 15°C 

after 2 months. Standard error bars with different letters are significantly 

different according to LSD at the 5% level. 

Figure 5.1: General embryo excision procedure for Romulea seeds. An outer 

view, as one would view it through a stereo microscope, as well as a view 

relative to the embryo is provided so that the importance of the placing of the 

incisions can be seen. Step 1 is viewed from the top, Step 2 is a side view, 

Steps 3 and 5 are bottom views 90° to the incision made in Step 2. Step 4 is a 

side view. 

Figure 5.2: Effect of kinetin concentration on shoot production of Romulea 

diversiformis embryos after 2 months. Error bars indicate standard error of the 

mean. 


flava embryos after 2 months. Error bars indicate standard error of the mean. 

Letters indicates significance differences between treatments according to 

Duncan’s multiple range test. 


minutiflora embryos after 2 months. Error bars indicate standard error of the 

mean. 


monadelpha embryos after 2 months. Error bars indicate standard error of the 

mean. 


sabulosa after 2 months. Error bars indicate standard error of the mean. 

Letters shows significance differences between treatments according to 

Tukey’s HSD test. 

Figure 5.7: Visual observations of Romulea sabulosa cultures. Cultures 

including both kinetin and 2,4-D appears to exhibit embryo-like structures. 

Figure 5.8: The effect of three different concentrations of kinetin and mTR 

either with or without 0.5 NAA on shoot production of Romulea leipoldtii 

seedling hypocotyls and embryos. Error bars indicate standard errors of the 

means. Letters show significant differences between treatments according to 

Duncan’s multiple range test. 

Figure 5.9: Effect of three different concentrations of five cytokinins on 

multiplication of Romulea sabulosa shoots after 2 months. Error bars indicate 

standard error of the mean. Letters shows significant differences between 

treatments according to Duncan’s multiple range test. 

Figure 6.1: An in vitro formed flower of Romulea minutiflora observed in a test 

tube placed at 20°C on a medium with 9% sucrose. 

108 

110 

118 

124 

124 

125 

125 

129 

130 

132 

133 

145 

xvi

Figure 6.2: Corms of Romulea sabulosa growing in a modified plastic 

container with vermiculite after 2 months. Bar = 20 mm. 


Figure 7.1. Showing eight species used in propagation experiments arranged 

from the largest to the smallest growth form. From the left they are Romulea 

minutiflora (A), R. camerooniana (B), R. diversiformis (C), R. rosea (D), R. 

flava (E), R. leipoldtii (F), R. monadelpha (G) and R. sabulosa (H). Modified 

from DE VOS (1972) and photographs taken by Dr. John C. Manning. 

Horizontal bar = 50 mm. 

146 

151 

xvii

List of Tables 

Table 2.1: Names of the soil separates and the particle diameters which define 

them (Modified from DONAHUE et al. (1983)). 

Table 2.2: Classification of mineral elements into macro- and micronutients 

(Modified from (MARSCHNER, 1995)). 

Table 2.3: Organic seed endogenous and exogenous dormancy types 

(Modified from BASKIN & BASKIN (1998)). 

Table 2.4: Topographic stain evaluation classes for the TTC test (LEADEM, 

1984). 

Table 2.5: The standard MURASHIGE & SKOOG (1962) formula. 

Table 2.6: Example of a matrix to establish optimal auxin to cytokinin ratios 

and their concentrations, where the rows represent auxin levels and the 

columns represent the cytokinin levels (Modified from Kyte and Kleyn (1996). 

Table 2.7: Explant sources and PGR's used by various authors for direct shoot 

or meristimoid organogenesis in genera of Iridaceae. Where the 

concentrations of PGR's are not mentioned, the study included multiple 

species within the genus, each reacted differently to various concentrations. A 

question mark indicates that the specific parameter is not included in the 

described publication. The genera are grouped phylogenetically, with vertical 

text on the right showing classification. 

Table 2.8: Corm induction treatments for various genera in Iridaceae. Details 

on the media modifications, temperature and the hours of light (Photoperoid) 

during corm induction is included. The peroid it took for corms to form is also 

given in months. The genera are grouped phylogenetically, with vertical text on 

the right showing classification. A question mark indicates that the specific 

parameter is not included in the described publication. 

Table 3.1: Analysis results for two soil samples from the Nieuwoudtville 

Wildflower Reserve (19° 8’ E, 31° 24’ S). 

Table 4.1: Seed viability tests of different Romulea species. 

Table 4.2: Effect of different treatments on seed germination of four Romulea 

species. Asterisk (*) indicates seed germination under 16 h photoperiod at 20 

± 0.5°C. The number sign (#) indicates that the seeds initiated germination 

during stratification. 

Table 5.1: Effect of kinetin and 2,4-D on excised embryos of Romulea 

sabulosa. Mean values in a column followed by different letters that indicates 

significance differences between treatments according to Duncan’s multiple 

range test (P 0.05). S = swelling of embryo; SR = swelling of embryo with 

rooting; SSI = swelling of embryo with shoot initials; SRF = shoot and root 

37 

40 

55 

59 

68 

74 

87 

90 

96 

103 

107 

xviii

formation; SC = shoot cluster; SCR = shoot cluster with roots; CSCI = callus 

with shoot cluster initials; CIS = corm-like structure (< 10 mm); CAI = callus 

appearing incompetent; CPE = callus with potential embryogenesis; PDE = 

potential direct embryogenesis; CSGR = cultures showed growth response. 

Potential embryogenesis refers to cultures that appeared to develop embryolike 

structures (Figure 5.7). 

Table 6.1: The effect of different temperatures and media composition on the 

in vitro formation and growth of Romulea minutiflora corms. Data shows the 

means ± the standard error. Letters indicates significant differences between 


Table 6.2: Percentage corm induction for Romulea minutiflora shoots cultured 

on medium supplemented with growth retardants. 

Table 6.3: The effect of different temperatures and media composition on the 

in vitro formation and growth of Romulea sabulosa corms. Data shows the 

means ± the standard error. Letters indicate significant differences between 


Table 6.4: Cultures with multiple corm formation for Romulea sabulosa. This 

shows the percentage of corm formation in cultures in which corm formation 

observed (Total cultures with corms) and the average number of corms 

produced in instances of multiple corm formation. 

List of Tables 

128 

142 

143 

144 

144 

xix

List of Abbreviations 

2,4-D 2,4-Dichlorophenoxyacetic acid 

2-iP N6(2-isopentenyl)-adenine 

ABA Abscisic acid 

BA 6-Benzyl-aminopurine 

Butenolide 3-methyl-2H-furo[2,3-c]pyran-2-one 

CaCl2 Calcium chloride 

CaCO3 Calcium carbonate 

GA Gibberellic acid 

HCl Hydrochloric acid 

IAA Indole-3-acetic acid 

IBA Indole-3-butryric acid 

Kinetin 6-Furfurylaminopurine 

KNO3 Potassium nitrate 

KOH Potassium hydroxide 

MemT 6-(3-methoxybenzylamino)purine 

MemTR 6-(3-methoxybenzylamino)-9-b-D-ribofuranosylpurine 

MGT Mean germination time 

MPa Mega Pascal 

MS Murashige and Skoog 

mT 6-(3-hydroxybenzylamino)purine 

mTR 6-(3-hydroxybenzylamino)-9-b-D-ribofuranosylpurine 

NAA Naphthaleneacetic acid 

NADPH Nicotinamide Adenine Dinucleotide Phosphate 

NaOH Sodium hydroxide 

nm nanometer 

PGR Plant Growth Regulator 

Thidiazuron N-phenyl-N-1,2,3,-thidiazol-5-ylurea 

TMS Table Mountain Sandstone 

TTC 2,3,5-triphenyltetrazolium chloride 

Zeatin Trans-6-(4-hydroxyl-3-methylbut-2enyl) aminopurine 

µM Micromole 

µm Micrometer 

xx

Namakwaland 

Die wêreld lê oop en kaal in die snikhete son 

hittegolwe bewe op die horison 

dit maak skimme op die grens 

die skaap soek koelte onder pens. 

The world lies open and naked in the hot sun 

heat waves quiver on the horizon 

it makes shimmers on the border 

the sheep searches for shade under paunch 

Die koggelmander skarrel stywebeen 

oor skroeiende granietklip heen. 

Geen koeltebome hier – die aarde is plat – 

Skilpad soek maar bossie, ander graaf maar gat 

The lizard scurries stiff-legged 

over scorching granite rock 

No shade trees here – the earth is flat – 

Tortoise looks for a small bush, others dig a hole 

Die klippe lê asof soos kaiings uitgebraai 

maar onder – deur die sandjies toegewaai – 

lê fyne saad gesaai, 

van moederplant al lank gespeen 

wagtend – op die reën. 

The rocks lie as if they were pieces of roasted crackling 

But underneath – covered by windblown sand – 

fine seeds are sown 

from parent plant weaned for long 

in waiting - for the rain . 

xxi

Die vlakhaas spits sy oor 

hy het die donderweer gehoor 

skilpad loer ook uit sy dop. 

Wildsbok lig en draai sy kop – 

sy neusgat vleuel soos hy die reënlug snuif 

daar gaan ñ rilling deur sy lyf 

sy stert staan kuif. 

The rabbit lifts his ear 

he has heard the thunder 

tortoise also peaks out his shell 

The wild buck lifts and turns his head - 

his nose flares as he smells the rain-filled air 

a shiver goes through his body 

his tail stands rigid 

Die donderweerswolke maak ñ donker sluier 

en die voorwind dwarrel en kuier 

hier en daar, van wie weet waar 

kielie ñ graspol en ritsel ñ blaar. 

Die eerste druppels val met sware plof 

in die dorre verpoeierde stof – 

dit maak ñ wasem op die klip 

en laat die sandtjies dans en wip. 

Water drup van die klip se rand 

in die dorstige rooi-rooi sand. 

The thunderclouds make a dark slur 

and the fore-wind tornadoes and visits 

here and there, from who knows where 

a grass is tickling and a leaf scurrying. 

The first drops fall with heavily, explosively 

in the dull powdered dust - 

it makes a haze on the rock 

and makes the sand particles dance and whip 

Water drips from the rock’s edge 

in the thirsty red-red sand. 

Namakwaland 

xxii

Dit lyk kompleet of die wêreld wil sing 

oor die genade wat weer uitkoms bring, 

want gou is die wêreld met blomme verfraai 

wat jubel van kleur as die wind daaroor waai 

It appears truly that the world wants to sing 

about mercifulness that again brings salvation, 

because instantaneously the world is bedazzled with flowers 

that rejoice with colour as wind blows over them 

Het iemand met ñ towerstaf 

ongesiens hier deurgedraf? 

Lap-lap lê dit aanmekaar 

ñ mooi gesig voorwaar! 

Gousblom en vergeet-my-nietjie 

met blou en pers so bietjie-bietjie. 

Did someone with a magic wand 

run through here unseen? 

Patch-patch it lies continuously 

a pretty face for sure! 

Arctotis hirsuta and Anchusa capensis 

with blue and purple little-little 

Slanguinjtjie weet nie wat haar noop 

maar stoot haar blommetjie skaam-skaam oop 

sy mag mos ook in die vreugte deel 

met skuterwit en spikkelgeel. 

Morea serpentina does not know what and where 

but pushes open her flower shy-shy 

she may also have her share of the happiness 

Geel en goud, bankvas aanmekaar 

met blouselblou, ñ bietjie hier, ñ bietjie daar. 

Yellow and gold, tightly packed together 

with blossom-blue, a little here, a little there 

ñ Wuiwend blommeparadys 

om sekerlik die Heer te prys. 

A lush flower paradise 

to surely praise the Lord 

Namakwaland 

xxiii

As die mens dan ook so sy dankbaarheid betoon, 

sal Hy sekerlik ook nader aan ons woon. 

If man then also can show his gratitude like this, 

then He will surely also live closer to us 

A poem by my late father, Pierre André Swart Senior; to whom this thesis is 

dedicated (English translation in grey bolded text). 

Namakwaland 

xxiv

1 Introduction 

1.1 PROPAGATION OF ROMULEA SPECIES FOR HORTICULTURAL AND 

CONSERVATION PURPOSES 

Romulea is a genus with many species of potential horticultural value. The fast 

growth, attractive growth forms, regular flowering and diverse flower variation with 

many aesthetically pleasing colours, makes species of this genus prime candidates 

for commercialisation as miniature potted plants and cut flowers (MANNING & 

GOLDBLATT, 1996; 1997; NIEDERWIESER et al., 2002). 

The Iridaceae is one of the most horticulturally important families of monocotyledons. 

Most of the cultivated ornamental species indigenous to South Africa have come 

from this family (COETZEE et al., 1999; NIEDERWIESER et al., 2002; REINTEN & 

COETZEE, 2002). The two genera of Iridaceae most in demand by the world market 

as floricultural crops are Gladiolus and Freesia (COETZEE et al., 1999). The 

production of cut flowers of Gladiolus and Freesia is a million dollar industry in many 

parts of the world (GOLDBLATT, 1991). These two genera are placed in the same 

tribe, Ixieae, as the genus Romulea (GOLDBLATT, 1990). 

The name Romulea was borrowed from the city of Rome, in vicinity of which the 

genus was first described by Maratti in a small taxonomic study published in 1772. 

He proposed that this species was distinct from Crocus, Colchicum, Sisyrinchium, 

Bulbocodium and Ixia (DE VOS, 1972). 

According to MANNING & GOLDBLATT (2001) there are approximately 90 species 

of Romulea. These species are found in sub-Saharan Africa, the Mediterranean 

basin, the Canary Islands, the Azores, and southern Europe (DE VOS, 1972; 

MANNING & GOLDBLATT, 2001). This attractive genus of the Iridaceae has its 

centre of diversity in the winter-rainfall zone of South Africa where 73 species are 

now recognized (MANNING & GOLDBLATT, 2001). Viewing the flowering plants in 

this area is an important tourist attraction. It attracts international tourists including 

world renowned botanists and nature lovers. Many South Africans also travel across 

the country each year to immerse themselves in the beauty of this floral spectacle, 

1

Introduction 

where a great number of species of Romulea can be seen. Within the summer- 

rainfall zone of southern Africa, the species is restricted to upland and montane 

habitats (MANNING & GOLDBLATT, 2001). Species belonging to Romulea are 

deciduous perennial geophytes and the tunicated corms of these plants enable them 

to survive the dry season (DE VOS, 1972; MANNING & GOLDBLATT, 2001). At the 

start of the growing season, a group of adventitious roots are first formed near the 

base of the corm, after which the uppermost axillary bud develops into an 

inflorescence stem (DE VOS, 1972). 

According to HILTON-TAYLOR (1996) there are 18 rare, 10 vulnerable and 2 

extinct species in the genus Romulea. RAIMONDO et al (2009) however lists this 

genus as having only 4 rare, 4 near threatened, 23 vulnerable, 7 endangered and 3 

critically endangered species. In their book, RAIMONDO et al (2009) displays a 

photograph of the vulnerable Romulea sabulosa on the cover. This species was used 

in this study. Despite the fragile conservation status of many species in this genus, 

the area which hosts its centre of diversity is also under threat from climate change. 

The longer periods and higher intensity of drought in the Cape Floral Region is likely 

to have a large negative impact on the endemic flora (WEST, 2009). 

This study will form the groundwork for the commercialisation and conservation of 

this genus, as there has been no extensive work done on its ecophysiology and 

propagation. 

Micropropagation is an important tool for ornamental plant culture and breeding, 

which has been applied to almost all commercial geophytes (ZIV, 1997). It enables 

high propagation rates, which is especially useful for the commercialisation of new 

species (LILIEN-KIPNIS & KOCHBA, 1987; PIERIK, 1997). In many instances it has 

also been shown that micropropagation can play a vital role in plant conservation, 

especially when combined with methods such as cryopreservation (WOCHOCK, 

1981; SARASAN et al., 2006; SHIBLI et al., 2006; WITHERS, 2008). 

2

1.2 AIMS AND HYPOTHESES 

Introduction 

The general aim of this study was to investigate the conditions that promote growth 

and development in a number of Romulea species both ex vitro and in vitro to aid its 

commercialisation and conservation. In more detail, the aims of this study were: 

• To investigate the environmental factors that influence the development and 

growth of R. sabulosa and R. leipoldtii in their natural habitat and to replicate 

these conditions for ex vitro growth; 

• To investigate the germination physiology of R. camerooniana, R. 

diversiformis, R. flava, R. leipoldtii, R. minutiflora, R. monadelpha, R. rosea 

and R. sabulosa and to improve seed germination in some of these species 

that show low germination; 

• To obtain suitable protocols to initiate cultures for the micropropagation of R. 

diversiformis, R. flava, R. leipoldtii, R. monadelpha, R. minutiflora and R. 

sabulosa; 

• To establish suitable protocols for shoot multiplication for R. leipoldtii, R. 

minutiflora and R. sabulosa; 

• To investigate whether embryogenesis readily occurs in the presence of 2,4-D; 

• To establish protocols for in vitro corm production and ex vitro corm 

germination for R. leipoldtii, R. minutiflora and R. sabulosa; 

• To establish an in vitro flowering protocol for R. minutiflora; and 

• To promote the commercialisation of Romulea species 

It was expected that there will be a correlation between geographical distribution and 

suitable ex vitro and in vitro stimuli. The suitable culture conditions were expected to 

be similar to that of Crocus species due to their close relationship with this genus. 

3

Introduction 

The subgenera Romulea and Spatulanthus were expected to have differentially 

suitable conditions for ex vitro and in vitro development and growth. 

1.3 GENERAL OVERVIEW OF THESIS CONTENT 

Chapter 2 is a review of literature available on aspects relative to this study. It firstly 

covers the distribution, morphology, life-cycle, habitat and conservation status of 

species in the genus Romulea. It then discusses other studies performed on 

ecophysiology and propagation of this genus with descriptions of phylogeny and 

taxonomy. It further reviews the ecophysiological techniques of soil sampling and 

analysis. A review of seed physiology and techniques applicable to this study is 

included in this chapter. It also gives a review on micropropagation in general, 

discusses some in vitro techniques applicable to the study, placing emphasis on 

explant selection, culture initiation and multiplication, embryogenesis, in vitro corm 

formation, in vitro flowering and ex vitro acclimatization. A summary of the 

micropropagation of species in the family Iridaceae is included. 

In Chapter 3 the habitat of some Romulea species is investigated further through 

ecophysiological techniques of soil sampling and analysis. 

Chapter 4 is an examination of the germination physiology of some Romulea 

species. This was done firstly by examining the physical properties and viability of the 

seeds, and then investigating the effect of an array of physical and chemical stimuli 

on germination. The physical properties of the seeds; imbibition rate, moisture 

content and viability of seeds were determined. The seed coat and micropylar 

regions were examined using scanning electron microscopy. To test for suitable 

stimuli for germination, the effect of temperature and light, cold and warm 

stratification, acid and sand paper scarification, plant growth promoting substances, 

deficiency of nitrogen, phosphorous and potassium, and different light spectra 

(phytochromes) on germination were examined. 

Chapter 5 is an examination of the suitability of various explant types and media 

supplementations for culture initiation. Two explant types were used; seedling organs 

and embryos. It also investigates the effect of various physical and chemical 

4

Introduction 

parameters on shoot multiplication and describes some cultures that appeared 

embryogenic. 

Chapter 6 is a report on the effect of various physical and chemical parameters on in 

vitro corm formation and ex vitro acclimatization and growth. It includes a description 

of an incident of in vitro flowering and some experiments conducted in an attempt to 

replicate these conditions and further stimulate in vitro flowering. 

In Chapter 7 the attributes of various Romulea species is considered and their 

suitability for commercialization is discussed. 

5

2 Literature review 

Leef van daad en woord dan so gepas, 

Dat jy nooit wens dat môre gister was 

Live of word and deed then in such a becoming way, 

That you never shall wish tomorrow was yesterday 

Pierre André Swart Senior 

2.1 MORPHOLOGY, DISTRIBUTION AND HABITAT 

There are approximately 90 species of Romulea (MANNING & GOLDBLATT, 

2001). These species are mainly confined to sub-Saharan Africa and the 

Mediterranean (DE VOS, 1972; MANNING & GOLDBLATT, 2001). Twelve to 15 

species occur in the Mediterranean basin, Canary Islands, the Azores, and 

southern Europe (MANNING & GOLDBLATT, 2001). The remaining species 

occur in sub-Saharan African, which includes the Arabian Peninsula and 

Socotra (MANNING & GOLDBLATT, 2001). It is reported that 3 species occur in 

tropical Africa and 2 are endemic in East Africa and the Peninsula (MANNING & 

GOLDBLATT, 2001). The genus has its centre of diversity in the winter-rainfall 

zone of southern Africa; here 73 species are now recognized (MANNING & 

GOLDBLATT, 2001). The distribution of 7 species used in the propagation 

experiments are shown in Figure 2.1. Within the summer-rainfall zone of 

southern Africa the species are restricted to upland and montane habitats 

(MANNING & GOLDBLATT, 2001). Those within the winter rainfall part of 

southern Africa occur from sea level to high altitudes, being especially common 

in medium to high altitudes (MANNING & GOLDBLATT, 2001). Winter-rainfall 

species generally flower during the spring (August to September, with a few in 

May and June) and summer-rainfall species flower from September to February 

(MANNING & GOLDBLATT, 2001) as shown in Figures 2.2 to 2.4. 

6

Figure 2.1: Map showing the distribution of seven of the species used in propagation experiments. The inset of the globe in the top right 

corner indicates the location of this map on the African continent with a rectangle. Modified from DE VOS (1972; 1983). 

Literature review 

7


Figure 2.2: Life cycle of Romulea sabulosa, a species endemic to the winter-rainfall area of South Africa (Modified from ASCOUGH (2008); DE VOS 

(1972); and photographs taken by Dr. John C. Manning) 

8


Figure 2.3: Life cycle of Romulea monadelpha, another species endemic to the winter-rainfall area of South Africa (Modified from ASCOUGH 

(2008); DE VOS (1972); and photographs taken by Dr. John C. Manning). 

9


Figure 2.4: Life cycle of Romulea camerooniana, a species occurring in summer-rainfall regions of Africa (Modified from ASCOUGH (2008); DE VOS 

(1972) and photographs taken by Dr. John C. Manning). 

10


Species belonging to Romulea are deciduous perennial geophytes (DE VOS, 1972; 

MANNING & GOLDBLATT, 2001).The tunicated corms of these plants enable them 

to survive the dry season (DE VOS, 1972). At the start of the growing season a group 

of adventitious roots are first formed near the base of the corm after which the top 

axillary bud develops into a inflorescent stem (DE VOS, 1972). During growth the 

corm gradually shrinks and a new corm is formed, which remains dormant until the 

next season (DE VOS, 1972). 

Most of the species used in this study generally occur in seasonally moist or 

inundated open sandy or clay flats (MANNING & GOLDBLATT, 2001). The genus 

Romulea is not as substrate-specific as many other southern African Iridaceae and in 

Romulea sp. true edaphic endemics are rare (MANNING & GOLDBLATT, 2001). 

These edaphic endemics include most of the species endemic to the western Karoo, 

which is found only in fine-grained doleritic clay soil and R. barkerae, which is 

restricted to the coastal limestone deposits of the Saldanha district (MANNING & 

GOLDBLATT, 2001). 

The corms of species belonging to Romulea are described by as globose, bell- 

shaped or asymmetrical and woody (MANNING & GOLDBLATT, 2001). The corm 

consists of a few swollen, basal internodes of the axis covered by the tunics, which 

consists of persistent leaf bases (DE VOS, 1972). DE VOS (1972) noted that the 

adventitious roots of species belonging to this genus originate from a basal ridge or 

basal point on the corm in the form of a row or cluster that represents the ventral side 

of the rhizome from which the distinguishing corm of the species in the family 

Iridaceae evolved (DE VOS, 1972). When the plant is too high in the ground a 

contractile root may develop from the basal scar (DE VOS, 1972). This root is thicker 

than other adventitious roots (DE VOS, 1972). The new corm is most commonly 

obliquely attached to the old corm via the basal scar (DE VOS, 1972). This basal 

scar is not quite basal and is actually situated towards one side (DE VOS, 1972). A 

new corm develops when a basal internode of the axis develops into leafy shoots 

(DE VOS, 1972). The newly formed corms are still partially enclosed in the old corm 

tunics after formation (DE VOS, 1972). 

11


The plants are short stemmed or stemless. The flowering stems of Romulea species 

are also usually reduced and often subterranean. The flowers are each borne singly 

on a branch or peduncle (MANNING & GOLDBLATT, 2001). The leaves of Romulea 

are linear to filiform, with most species having two grooves on each surface. When 

the leaf is examined anatomically, it consists of a wide central rib separated from the 

smaller marginal ribs by wide to narrow longitudinal grooves. The stomata are 

located in these longitudinal grooves (MANNING & GOLDBLATT, 2001). 

The flowers of most species are very similar except for pigmentation, which is 

exceptionally variable. The colour array includes uniformly yellow to white, pink, 

orange, apricot, red, magenta, lilac and purple, with the cup usually being yellow (DE 

VOS, 1972; MANNING & GOLDBLATT, 2001). Dark markings commonly appear 

below the rim of the cup. The perianth is cup-shaped with a short perianth tube. The 

flower has six tepals, which are cupped below and spreads horizontally above. The 

floral cup includes the stamens which are adjoining and coherent. The style divides 

into three distinct style arms above mid-anther level (MANNING & GOLDBLATT, 

2001). The flowers are short lived and not suitable for picking (MANNING & 


2.2 SPECIES SPECIFIC MORPHOLOGY AND DISTRIBUTION 

According to MANNING & GOLDBLATT (2001) and DE VOS (1970a) the corm and 

its tunics appear to provide the most useful characteristics for identifying different 

species within the genus. In most species the corm develops a sharp lateral or basal 

ridge through intercalary growth of the tunics. The margins of the tunics along this 

fold consist of fine fibrils, forming a fibrous fringe (MANNING & GOLDBLATT, 2001). 

MANNING & GOLDBLATT (2001) classifies these species as belonging to the 

subgenus Romulea. 

The corms of some other species have a rounded or pointed base and lack a basal 

ridge. In this case the tunics split into several well-defined acuminate teeth that do 

not have a fibrous appearance (MANNING & GOLDBLATT, 2001). MANNING & 

12


GOLDBLATT (2001) classifies these species as belonging to the subgenus 

Spatalanthus. Manning and Goldblatt (2001) goes on to divide the genus into 

sections based on corm morphology, this is not discussed. 

Research was done on the character of flowers and growth form of a number of 

Romulea species found in South Africa. Descriptive localities, habitats, identifying 

features and subgenus classification of these species used in propagation 

experiments are also discussed. 

2.2.1 Romulea amoena 

The flowers of R. amoena are deep rose-pink to carmine-red with purple-black 

blotches (DE VOS, 1983; MANNING & GOLDBLATT, 1997; MANNING & 

GOLDBLATT, 2001). These blotches are sometimes also replaced by stripes around 

the cream or yellow cup. The tepals are elliptic to oblanceolate (MANNING & 

GOLDBLATT, 2001). The bracts are green or sometimes reddish (DE VOS, 1983). 

The outer bracts have narrow or inconspicuous membranous margins while the inner 

bracts have wide and colourless or brown-streaked margins. It has 1 to 2 flowers 

which can be seen in August (DE VOS, 1972; MANNING & GOLDBLATT, 1997). 

Plants of R. amoena are between 50 and 300 mm in height (DE VOS, 1983; 

MANNING & GOLDBLATT, 2001). Its stem is subterranean or reaches 100 mm 

above the ground. These plants have 3 to 4 leaves that are usually all basal in origin 

(MANNING & GOLDBLATT, 2001). 

R. amoena occurs in sandy soils and is mostly found in rocky places. It is indigenous 

to the Bokkeveld mountains south of Nieuwoudtville (MANNING & GOLDBLATT, 

2001). MANNING & GOLDBLATT (2001) has placed this species in the subgenus 

Romulea. 

2.2.2 Romulea autumnalis 

The flowers are pink or magenta-pink to white with a yellow to orange cup. The 

tepals are elliptic and the fruiting peduncles are erect. The bracts are green or 

greenish with the outer bracts having narrow membranous margins while the inner 

bracts have wide colourless margins (DE VOS, 1983; MANNING & GOLDBLATT, 

2001). R. autumnalis flowers from April to July during which the plant develops 1 to 3 

13


flowers (DE VOS, 1972; MANNING & GOLDBLATT, 2001). The plants of this 

species grow 150 to 350 mm in height with subterranean stems. The plant has 3 to 4 

leaves which are basal and thread like or filiform to compressed cylindrically (DE 

VOS, 1983; MANNING & GOLDBLATT, 2001). 

This species is found in Eastern Cape from Grahamstown towards Kariga where it 

occurs on grassy flats or mountain slopes (DE VOS, 1983; MANNING & 

GOLDBLATT, 2001). It is closely allied with R. camerooniana, but can be 

distinguished from R. camerooniana by its short stamens and style which do not 

reach the middle of the perianth, as opposed the stamens and styles of R. 

camerooniana, which do reach the floral cup (MANNING & GOLDBLATT, 2001). This 

means that the stamens of R. autumnalis are included in the floral cup (MANNING & 

GOLDBLATT, 2001). MANNING & GOLDBLATT (2001) places this species in the 


2.2.3 Romulea camerooniana 

The flowers are magenta or pink to white and the cup is yellow. The tepals are elliptic 

(MANNING & GOLDBLATT, 2001). The outer bracts have narrow or inconspicuous 

membranous margins. The inner bracts also have narrow and colourless 

membranous margins (MANNING & GOLDBLATT, 2001). R. camerooniana mostly 

flowers from December to April (BURROWS & WILLIS, 2005). The plants are 

normally 80 to 200 mm in height with a stem which is subterranean. There are 2 to 6 

filiform leaves per plant which are all basal (MANNING & GOLDBLATT, 2001). 

R. camerooniana occurs in rocky or grassy highlands. In these habitats their 

distribution extends from the Drakensberg of the Eastern Cape, South Africa to 

Kenya, Sudan and Southern Ethiopia. Outlying populations also occur in the 

Cameroon in west Africa (MANNING & GOLDBLATT, 2001). MANNING & 

GOLDBLATT (2001) places this species in the subgenus Romulea. 

2.2.4 Romulea citrina 

The flowers are lemon-yellow and unscented with tepals that are elliptic and between 

20 and 32 mm long (DE VOS, 1983; MANNING & GOLDBLATT, 2001). The fruiting 

peduncles are at first curved and later suberect. The outer bracts have narrow 

14


membranous margins while the inner bracts are slightly shorter with broad, brown 

streaked, membranous margins (DE VOS, 1983; MANNING & GOLDBLATT, 2001). 

R. citrina flowers from August to September with 1 to 4 flowers (DE VOS, 1972; 

MANNING & GOLDBLATT, 2001). These plants reach a height of 80 to 350 mm, and 

sometimes reach up to 450 mm. The stem is subterranean or reaches 20 mm above 

the ground. The plant has 3 to 4 filiform leaves of which the lower two are basal (DE 

VOS, 1983; MANNING & GOLDBLATT, 2001). The leaves are narrowly grooved with 

4 grooves. They are compressed cylindrically and curve outward (MANNING & 


This species occurs in wet sites with sandy or stony ground in Namaqualand where it 

is common in the Kamiesberg area (DE VOS, 1983; MANNING & GOLDBLATT, 

2001). lt also occurs at lower elevations and is found around Grootvlei, west of 

Kamieskroon (MANNING & GOLDBLATT, 2001). MANNING & GOLDBLATT (2001) 

places this species in the subgenus Romulea. 

2.2.5 Romulea cruciata 

Flowers are magenta-pink to lilac-pink in colour. There are dark blotches around a 

dark or golden yellow to orange cup (DE VOS, 1983; MANNING & GOLDBLATT, 

1996; MANNING & GOLDBLATT, 2001). The flower is unscented with elliptic to 

oblanceolate tepals. The fruiting peduncles remain erect or spread slightly. The 

bracts are greenish or purplish red with the outer bracts having narrow inconspicuous 

membranous margins and the inner bracts submembranous which are commonly 

brown-flecked (MANNING & GOLDBLATT, 2001). Its flowering period is between 

July and September during which a plant may have 2 to 4 flowers (DE VOS, 1972; 

MANNING & GOLDBLATT, 2001). These plants are 150 to 400 mm in height with a 

totally subterranean stem (DE VOS, 1972; MANNING & GOLDBLATT, 2001). The 

plant has 2 to 8 filiform leaves which are all basal (MANNING & GOLDBLATT, 2001). 

R. cruciata is most common in the south western Cape from Nieuwoudtville to 

Riversdale (MANNING & GOLDBLATT, 1996; MANNING & GOLDBLATT, 2001). Its 

distribution spreads from the Bokkeveld Mountains of the Northern Cape Province in 

the north to the Gourits river in the Western Cape Province in the east where it is 

15


often found on clay or granitic soils in renosterveld (MANNING & GOLDBLATT, 

2001). On the 7 th of September 2006 R. cruciata was observed in renosterveld 

outside Malmesbury by a group from Naturetrek (PONTING, 2006). MANNING & 

GOLDBLATT (2001) place this species in the subgenus Spatalanthus. 

2.2.6 Romulea diversiformis 

The flowers are buttercup-yellow and unscented (DE VOS, 1983; MANNING & 

GOLDBLATT, 1997; MANNING & GOLDBLATT, 2001). Tepals are obovate with the 

inner tepals broader than the outer tepals (MANNING & GOLDBLATT, 1997; 

MANNING & GOLDBLATT, 2001). The fruiting peduncles are bent (MANNING & 

GOLDBLATT, 2001). Bracts are green to greenish (DE VOS, 1983). The outer bracts 

have narrow white membranous margins and apices while the inner bracts have 

wider membranous margins. R. diversiformis flowers (1 or more flower per plant) 

from August to September (DE VOS, 1972; MANNING & GOLDBLATT, 1997). These 

plants are 80 to 200 mm in height and are classed as stemless geophytes. It has 6 to 

10 filiform leaves which are all basal (DE VOS, 1983; MANNING & GOLDBLATT, 

2001). 

The plants occur in moist or waterlogged dolerite and clay in the Western Karoo and 

Roggeveld of South Africa (MANNING & GOLDBLATT, 1997; MANNING & 

GOLDBLATT, 2001). MANNING & GOLDBLATT (2001) place this species in the 

subgenus Spatalanthus. 

2.2.7 Romulea flava 

Flowers are white or yellow and rarely flowers are also blue, blue-violet or pinkish 

(DE VOS, 1983; MANNING & GOLDBLATT, 1996; MANNING & GOLDBLATT, 

2001). The flowers have a yellow cup (MANNING & GOLDBLATT, 1996; MANNING 

& GOLDBLATT, 2001). The white flowers of R. flava are usually scented (MANNING 

& GOLDBLATT, 2001). The tepals are oblanceolate and the outer tepals are 

uniformly green on the abaxial side. The fruiting peduncles are recurved and later 

erect (MANNING & GOLDBLATT, 2001). Outer bracts have narrow or inconspicuous 

membranous margins while inner bracts are submembranous or membranous and 

are often brown streaked (DE VOS, 1983; MANNING & GOLDBLATT, 2001). These 

plants flower from June to September and each plant generally has 1 to 4 flowers 

16



2001). They are 50 to 550 mm in height with a stem that is subterranean or reaches 

300 mm above the ground (DE VOS, 1983; MANNING & GOLDBLATT, 2001). The 

plant has 3 to 4 leaves of which one is a basal leaf. Leaves are narrowly or widely 

grooved with 4 grooves and are sometimes minutely ciliate or filiform (DE VOS, 1983; 

MANNING & GOLDBLATT, 2001). 

R. flava populations are widespread in the southern African winter-rainfall zone 

(MANNING & GOLDBLATT, 2001). These plants grow in sandy or clay soils from 

Namaqualand in the north to Humansdorp in the southeast where it occurs in fynbos 

and renosterveld (MANNING & GOLDBLATT, 2001). MANNING & GOLDBLATT 

(2001) places this species in the subgenus Romulea. 

2.2.8 Romulea leipoldtii 

Flowers are white to cream with a yellow cup and sweetly scented (DE VOS, 1983; 

MANNING & GOLDBLATT, 2001). Tepals are elliptic and 18 to 35 mm long. 

Filaments are 5 to 8 mm long and anthers are 5 to 8 mm long. Fruiting peduncles are 

bent and later erect (MANNING & GOLDBLATT, 2001). Outer bracts are green with 

inconspicuous membranous margins. Inner bracts have colourless or brown speckled 

membranous margins (DE VOS, 1983; MANNING & GOLDBLATT, 2001). R. 

leipoldtii flowers from September to October (DE VOS, 1972; MANNING & 

GOLDBLATT, 2001). These plants usually have 4 to 6 flowers or more and are 100 

to 300 mm in height, with some plants reaching up to 600 mm (DE VOS, 1972; 

MANNING & GOLDBLATT, 2001). The stem reaches 50 to 350 mm above ground 

(DE VOS, 1983; MANNING & GOLDBLATT, 2001). The plant has 4 to 6 leaves of 

which the lower 2 are basal. These leaves are grooved narrowly with 4 grooves 


R. leipoldtii occurs from the Bokkeveld Mountains in the Northern Cape Province in 

the north to Klipheuwel near Malmesbury in Western Cape Province in the south 

where it is found growing in damp sandy soil (DE VOS, 1983; MANNING & 

GOLDBLATT, 2001). This species is closely allied with R. tabularis (MANNING & 

GOLDBLATT, 2001). The main difference is the larger, bicoloured, cream to white 

17


flowers of R. leipoldtii which has a dark yellow to orange centre (MANNING & 

GOLDBLATT, 2001). MANNING & GOLDBLATT (2001) place this species in the 


2.2.9 Romulea minutiflora 

The small flower of this species is pale mauve with a yellowish cup. The tepals are 4 

to 9 mm long and elliptic (MANNING & GOLDBLATT, 2001). The fruiting peduncles 

of this species are curved and later erect. The outer bracts have narrow margins 

which are frequently brown-spotted. Inner bracts can be observed to also have these 

brown-spotted margins (sometimes submembranous) (MANNING & GOLDBLATT, 

2001). Flowering occurs from July to September (DE VOS, 1972; MANNING & 

GOLDBLATT, 2001). These plants have been observed to have 1 to 4 flowers and to 

reach 60 to 200 mm in height (DE VOS, 1972; MANNING & GOLDBLATT, 2001). 

There are several basal leaves present. These measure between 0.5 and 1.5 mm in 

diameter and are narrowly 4-grooved (MANNING & GOLDBLATT, 2001). 

R. minutiflora is abundant throughout the South African winter-rainfall region. Its 

range extends from the Bokkeveld Mountains in the west to Grahamstown in the east 

(MANNING & GOLDBLATT, 2001). This species has been introduced into Australia 

(DE VOS, 1972). R. minutiflora is closely allied to R. sinispinosensis. Both these 

species have corms with a spade-shaped basal ridge (MANNING & GOLDBLATT, 

2001). R. minutiflora is however easily identifiable by its very small pale mauve or 

pink flowers of which the largest flower observed had tepals measuring a mere 9 mm 

in length (MANNING & GOLDBLATT, 2001). The tepals of R. sinispinosensis are 

white and between 10 and 12 mm in length (MANNING & GOLDBLATT, 2001). 

MANNING & GOLDBLATT (2001) place this species in the subgenus Romulea. 

2.2.10 Romulea monadelpha 

This attractive unscented species has large dark red flowers with black blotches at 

the edge of its creamy cup. The tepals of this species is 25 to 40 mm long and 

obovate-cuneate, resulting in an almost bell-shaped flower (DE VOS, 1970b). Its 

anther filaments are 3 to 4 mm long and oblong, adnate or fused into a column. The 

fruiting peduncles are curved. The upper portion of outer bracts is commonly one 

keeled and the inner bracts are two keeled. The outer and inner bracts both have 

18


brown membranous margins. Flowering is observed from August to September. The 

plants have 1 to 4 flowers (DE VOS, 1970b). DE VOS (1972) mentions that this is 

one of the most beautiful romuleas. The plants are 150 to 300 mm high with a 

subterranean stem. The plant has 3-5 filiform (1 to 2 mm in diameter) basal leaves 

which have 4 grooves (MANNING & GOLDBLATT, 2001). 

The habitat of R. monadelpha is restricted to the Northern Cape Province of South 

Africa. Here it occurs on dolerite clay in an area that starts in the proximity of 

Nieuwoudtville, extends south to the top of the Gannaga Pass near Middelpos and 

stretches along the Bokkeveld and Roggeveld Escarpments in the western Karoo 

(MANNING & GOLDBLATT, 2001). The flowers of the Gannaga Pass population has 

salmon pink flowers with large silvery grey and black markings around the cup 


It is interesting to note that a plant collector brought a few corms of R. monadelpha to 

England in 1825. Two of the corms germinated, developed and flowered in Colvill’s 

Nursery soon after planting. Unfortunately the locality of collection was not known 

and the only knowledge that existed for 139 years was drawings made during this 

discovery (DE VOS, 1970b). It appears that no herbarium specimens were made in 

this study. 

Then in 1964 on a search for R. sabulosa in the area of Calvinia, a species that was 

so similar to R. sabulosa was found, that it was mistaken for R. sabulosa (DE VOS, 

1970b). After being cultivated (no details on growth conditions given) in the 

Stellenbosch Nursery it was however noted that these had a fused filament column, a 

distinguishing feature of R. monadelpha (DE VOS, 1970b). 

In a more recent collection by MANNING & GOLDBLATT (2001) it was noted that it is 

more usual for the filaments to be merely adnate. They suggest identifying R. 

monadelpha by its short black filaments which are oblong, unlike the usually pale 

green slender, tapering filaments of R. sabulosa. These two species can also be 

distinguished by their fruiting peduncles (MANNING & GOLDBLATT, 2001). R. 

monadelpha typically has stout and semiterete peduncles with conspicuously 

19


flattened upper sides which are curved to the fruit. The fruiting peduncles of R. 

sabulosa, on the other hand, are commonly more slender and rounded in section and 

remain suberect in fruit. The peduncles of the latter species also remain suberect 

during fruiting. Apart from these morphological differences R. sabulosa is also 

restricted to light sandy clay soils near Nieuwoudtville whereas R. monadelpha is 

found on heavy, dolerite clay in several localities along the Bokkeveld and Roggeveld 

escarpments (MANNING & GOLDBLATT, 2001). MANNING & GOLDBLATT (2001) 

places this species in the subgenus Spatalanthus. 

2.2.11 Romulea pearsonii 

The flowers are lemon-yellow with elliptic tepals and the fruiting peduncles are 

suberect (DE VOS, 1983; MANNING & GOLDBLATT, 2001). Both the outer and the 

inner bracts are green. The outer bracts are firm and closely veined with narrow 

brown streaked membranous margins and an apex (DE VOS, 1983; MANNING & 

GOLDBLATT, 2001). The inner bracts have broad brown streaked membranous 

margins (MANNING & GOLDBLATT, 2001). The plant flowers in August to 

September with 1 to 3 flowers (DE VOS, 1972; MANNING & GOLDBLATT, 2001). 

Plants of this species have a height of 100 to 250 mm (DE VOS, 1983; MANNING & 

GOLDBLATT, 2001). The stem is completely subterranean or reaches 30 mm above 

ground (DE VOS, 1983; MANNING & GOLDBLATT, 2001). The plant has 3 to 4 

filiform leaves of which 2 are basal (MANNING & GOLDBLATT, 2001). 

R. pearsonii is restricted to higher elevations in central Namaqualand (MANNING & 

GOLDBLATT, 2001). Here it occurs from Grootvlei and the main Kamiesberg range 

and grows in sandy and granitic slopes and flats (MANNING & GOLDBLATT, 2001). 

MANNING & GOLDBLATT (2001) places this species in the subgenus Romulea. 

2.2.12 Romulea rosea 

The flowers are pink to magenta or white with a purplish zone around the yellow cup 

and are occasionally scented. The tepals are elliptic to oblanceolate and 4 to 6 mm 

long. The outer bracts have narrow membranous margins whereas the inner bracts 

have wide brownish membranous margins (MANNING & GOLDBLATT, 2001). R. 

rosea flowers from July to October and the plants have several flowers (DE VOS, 

1972; MANNING & GOLDBLATT, 2001). These plants reach 150 to 600 mm from the 

20


ground and have a subterranean stem (MANNING & GOLDBLATT, 2001). R. rosea 

has 3 to 6 leaves which are all basal. 

R. rosea is common and many varieties and forms are found (DE VOS, 1972). It 

occurs in a variety of habitats which include stony clay flats and slopes (MANNING & 

GOLDBLATT, 2001). It is found throughout the Cape region from the Bokkeveld 

range to Port Elizabeth. 

This species is a known invasive in other parts of the world, including Australia, New 

Zealand and USA (EDDY & SMITH, 1975; CROSSMAN et al., 2008; VAN KLEUNEN 

et al., 2008; FLEMING et al., 2009). Here it is regarded as a weed and numerous 

strategies can be found to eradicate this species due to its moderate toxicity to sheep 

and goats (EDDY & SMITH, 1975; SIMMONDS et al., 2000). MANNING & 

GOLDBLATT (2001) places this species in the subgenus Spatalanthus. 

2.2.13 Romulea sabulosa 

The flowers are currant or glossy red and rarely pink with black blotches at the edge 

of the creamy or greyish green cup (DE VOS, 1983; MANNING & GOLDBLATT, 

2001). The flowers are unscented with tepals that are obovate-cuneate. The fruiting 

peduncles are suberect (MANNING & GOLDBLATT, 2001). The outer bracts are 

usually keeled above with narrow, usually brown, membranous margins. The inner 

bracts are 2-keeled and usually also have brown membranous margins (MANNING & 

GOLDBLATT, 2001). The plants flower in July to September during which they may 

have 1 to 4 flowers (DE VOS, 1972; MANNING & GOLDBLATT, 2001). Plants are 

120 to 400 mm in length with a subterranean stem (MANNING & GOLDBLATT, 1997; 

MANNING & GOLDBLATT, 2001). The plant has 3 to 5 filiform leaves, all of which 

are basal (MANNING & GOLDBLATT, 2001). 

R. sabulosa is a local endemic that grows in renosterveld on clay and in the 

Bokkeveld Escarpment west of Nieuwoudtville on sandy soil (MANNING & 

GOLDBLATT, 1997; MANNING & GOLDBLATT, 2001). MANNING & GOLDBLATT 

(2001) places R. sabulosa in the subgenus Spatalanthus. 

21

2.2.16 Romulea tabularis 


Flowers are lavender blue to white or bluish-mauve. The flower has a yellow cup and 

the lower half of the tepals are yellow (DE VOS, 1983; MANNING & GOLDBLATT, 

2001). They are sometimes fragrant with tepals that are elliptic (MANNING & 

GOLDBLATT, 2001). The fruiting peduncles are arching and later erect. Outer bracts 

have inconspicuous membranous margins. The inner bract is submembranous with 

wide brown-speckled membranous margins (MANNING & GOLDBLATT, 2001). 

These plants flower from July to August with 2 to 4 flowers (DE VOS, 1972; 

MANNING & GOLDBLATT, 2001). Plants are 100 to 350 mm in height with a stem 

that reaches 100 to 350 mm above ground. They have 3 to 5 leaves of which 1 or 2 

are basal (DE VOS, 1983; MANNING & GOLDBLATT, 2001). 

R. tabularis occurs from northern Namaqualand to Cape Agulhas. Here it grows in 

wet, often waterlogged, sandy soils or limestone flats from Clanwilliam to Bredasdorp 


2001). Very closely allied with R. leipoldtii, R. tabularis however has smaller flowers 

and bicoloured tepals (MANNING & GOLDBLATT, 2001). MANNING & GOLDBLATT 

(2001) places this species in the subgenus Romulea. 

2.3 PHYLOGENY AND TAXONOMY 

The Iridaceae is a large family of petaloid monocotyledons. It includes over 1 630 

species in about 77 genera (GOLDBLATT, 1990). Within the Iridaceae there are the 

subfamilies of Isophysidoideae, Nivenoideae, Iridoideae and Ixioideae. Romulea 

belongs to the subfamily Ixioideae (GOLDBLATT, 1990). Ixioideae in turn contains 

the tribes Pilansiae, Watsonieae and Ixieae. Romulea is a member of Ixieae 

(GOLDBLATT, 1990). Ixioideae is centered in Southern Africa, but some species 

such as Gladiolus, Crocus and Romulea also occur in Eurasia. The subfamily 

comprises of over 860 species of which 760 belong to the tribe Ixieae (GOLDBLATT, 

1990). 

22


Within Ixieae there are the subtribes of Ixiinae, Tritoniinae, Gladiolinae, 

Radinosiphon, Hesperanthinae, Melasphaerula, Babianinae, Romuleinae, Freesiinae 

and Anapalinae. Romulea belongs to the subtribe of Romuleinae (GOLDBLATT, 

1990). The other members of this subtribe are Crocus and Syringodea 

(GOLDBLATT, 1990). 

Other genera in Ixieae includes Anomatheca, Babiana, Chasmanthe, Crocosmia, 

Crocus, Devia, Dierama, Duthieatrum, Freesia, Geissorhiza, Gladiolus, Hesperantha, 

Ixia, Melasphaetula, Radinosiphon, Shizostylis, Syringodea, Tritonia, Tritonopsis and 

Zygotritonia (GOLDBLATT, 1990). 

REEVES et al. (2001) showed the close association of the two genera Romulea and 

Crocus. It also confirms the classification of GOLDBLATT (1990) with molecular 

techniques. This classification was obtained by REEVES et al. (2001) by combining 

sequencing work in the Iridaceae done by other studies with their own sequencing 

data and grouping these in a combined parsimonious tree with bootstrap 

percentages and Fitch weights. 

Maratti did a small taxonomic study on Romulea species in 1772 (DE VOS, 1972). In 

this work he described Romulea after a species growing in the neighbourhood of 

Rome. He proposed that this species was distinct from Crocus, Colchicum, 

Sisyrinchium, Bulbocodium and Ixia (DE VOS, 1972). 

2.4 CONSERVATION STATUS 

As it is a moral duty of every scientist to be concerned about the conservation of 

species of which they are using reproductive material harvested from the wild, a 

section on the conservation status of this genus is therefore included. 

RAIMONDO et al. (2009) list Romulea as having fewer endangered species than 

previous studies (HILTON-TAYLOR, 1996). Although this is explained, it does not 

mention the status of one attractive species listed by HILTON-TAYLOR (1996) as 

23


extinct, R. papyracea. According to RAIMONDO et al. (2009) there are 3 critically 

endangered species, 7 endangered species, 23 vulnerable species, 4 non- 

threatened, 4 rare, 2 data deficient, and 59 species of least concern. 

Most of the species used in this study are classed by RAIMONDO et al. (2009) as 

species of least concern or species with a very low risk of extinction. Exceptions are 

R. pearsonii and R. sabulosa which are listed as being vulnerable to extinction. 

2.5 THE CLIMATE OF ROMULEA SPP. HABITATS 

The Namaqualand, where this genus has its centre of diversity, has a unique climate 

(DESMET, 2007). This is partially due to the fact that it is under the influence by two 

pronounced geographical rainfall gradients (DESMET, 2007). These are a latitudinal 

gradient of aridity and a longitudinal gradient of precipitation. The former decreases 

in precipitation towards the north into the Namib dessert and the latter brings 

precipitation from the winter-rainfall coastal areas and some summer-rainfall inland 

areas (DESMET, 2007). As a result, Namaqualand has highly reliable rainfall when 

compared to other arid regions with similar mean annual precipitation. Although 

rainfall in Namaqualand is low, it arrives between May and September almost every 

year. More than 60% of this rain is recorded during winter (DESMET, 2007). The cold 

Atlantic Ocean stabilises the climate of this region by preventing the intrusion of 

unstable air and by cooling this area with the aid of a pervasive south-westerly sea 

breeze (DESMET, 2007). 

The climatic conditions of regions where 8 of the species used in this study are 

commonly found are further described in more detail in the following paragraphs. 

Linking this data to data of flowering time and assuming that the flower will take a 

month to set seed gives a valuable insight into the temperature regime required for 

germination. Although only the above mentioned climatic conditions are discussed 

here, looking at the data a month before flowering also provides an estimation of 

what the most suitable temperature for further growth could be, whereas the climate 

at the time of flowering provides insight into possible physical stimuli for flowering. 

24


R. camerooniana has been known to occur in the Drakensberg region (MANNING & 

GOLDBLATT, 2001). The data in Figures 2.5 to 2.7 describes the climate of Royal 

National Park, a park situated within the Drakensberg range. This has been known to 

flower from May to February, although it mostly flowers from December to April 

(MANNING & GOLDBLATT, 2001; BURROWS & WILLIS, 2005), therefore it would 

be expected to set seed during May and July. 

In Figures 2.6 and 2.7 there is a large drop in average total monthly rainfall and a 

corresponding drop in average daily humidity for the month of April. These 

conditions, which are not suitable for germination, continue until September. During 

September, the average daily minimum and maximum temperatures (averaged for 

2004 to 2009) at the weather station at Royal National Park has been reported to be 

7.4±0.8°C and 24.5±0.8°C respectively. 

The climate of Calvinia, where R. diversiformis is found, is described in Figures 2.8 to 

2.10. R. diversiformis also occurs around the area of Sutherland, Fraserburg and 

Beaufort west (Figures 2.11 to 2.19). During the time in which this species is 

expected to set seed (October to November) there is an increase in average daily 

temperature and average total monthly rainfall and a decrease in average daily 

humidity. According to this data, seeds should germinate well with a regime of night 

temperatures between 0 and 15°C and day temperatures between 20 and 30°C. 

R. flava is widespread across a large part of the southern-African winter-rainfall zone, 

including areas near Nieuwoudtville (Figures 2.20 to 2.22). R. monadelpha and R. 

sabulosa and many other species in this genus also occur in this area (MANNING & 

GOLDBLATT, 2001). Here rainfall increases sharply in the month of October and 

average total monthly rain remains the same for November. The time in which these 

species are expected to set seed, October and November, appears to be a suitable 

time for seed germination in Nieuwoudtville, the rainfall also increases in December, 

providing moist conditions which are suitable for seedling establishment. Minimum 

and maximum temperatures are 8.2±0.3 and 25.6±0.4°C for October and 10.1±0.3 

and 27.5±0.4°C for November in Nieuwoudtville. 

25


R. leipoldtii occurs in areas near Malmesbury (Figures 2.23 to 2.25) (MANNING & 

GOLDBLATT, 2001). Here there is an increase in temperature from October to 

November with a corresponding increase in rainfall and decrease in humidity. In 

November the average daily temperature increases, but average total monthly rainfall 

and average daily relative humidity decreases. It is therefore more likely that these 

plants would set seed in October, when conditions are moist. This means that a 

minimum and maximum temperature of 9.3± 0.4 and 24.2±0.4°C should be suitable 

for germination. 

R. minutiflora is also widespread across a large part of the southern-African winter- 

rainfall zone, occurring as far east as Grahamstown (Figures 2.26 to 2.28) 

(MANNING & GOLDBLATT, 2001). During October the average daily minimum and 

maximum temperatures are 10.4±0.3 and 23.0±0.7°C, while the average daily 

minimum and maximum temperatures are 11.7±0.3 and 23.2±0.4°C during 

November. R. rosea is found throughout the Cape region and therefore any of these 

environments should be suitable for its germination. This is apparent when 

considering its distribution as illustrated in Figure 2.1. This species also has a long 

flowering time, lasting 4 months (July to October) (MANNING & GOLDBLATT, 2001). 

Estimation of conditions favourable for germination, such as those made for other 

species, is therefore unsuitable for this species. 

26


Figure 2.5: Royal National Park weather station (28° 57’ E, 28° 41’ S, 1392 m above sea level) 

average daily minimum and maximum monthly temperatures (Error bars indicate standard 



average total monthly daily rain (Error bars indicate standard error of the mean of last 5 

years). 


average daily relative humidity (Error bars indicate standard error of the mean of last 3 years). 

27


Figure 2.8: Calvinia (19° 56’ E, 31° 29’ S, 977 m above sea level) average daily minimum and 

maximum monthly temperatures (Error bars indicate standard error of the mean of last 5 

years). 

Figure 2.9: Calvinia weather station (19° 56’ E, 31° 29’ S, 977 m above sea level) average total 

monthly daily rain (Error bars indicate standard error of the mean of last 5 years). 

Figure 2.10: Calvinia weather station (19° 56’ E, 31° 29’ S, 977 m above sea level) average daily 

relative humidity (Error bars indicate standard error of the mean of last 5 years). 

28


Figure 2.11: Sutherland weather station (20° 4’ E, 32° 24’ S, 1458 m above sea level) average 

daily minimum and maximum monthly temperatures (Error bars indicate standard error of the 



total monthly daily rain (Error bars indicate standard error of the mean of last 5 years). 


daily relative humidity (Error bars indicate standard error of the mean of last 5 years). 

29


Figure 2.14: Fraserburg weather station (31° 55’ S 21° 30’ E, 1267 m above sea level) average 







30


Figure 2.17: Beaufort West weather station (22° 35’ E, 32° 21’ S, 899 m above sea level) average 







31

Figure 2.20: Nieuwoudtville weather station (19° 53’ E, 31° 21’ S, 731 m above sea level) 


average daily minimum and maximum monthly temperatures (Error bars indicate standard 



average total monthly daily rain (Error bars indicate standard error of the mean of last 5 years). 


average daily relative humidity (Error bars indicate standard error of the mean of last 3 years). 

32


Figure 2.23: Malmesbury weather station (18° 43’ E, 33° 28’ S, 108 m above sea level) average 







33


Figure 2.26: Grahamstown weather station (26° 30’ E, 33° 17’ S, 642 m above sea level) average 







34

2.6 SOIL SAMPLING AND ANALYSIS 


Soils are complex systems high in spatial variation. It universally consists of three 

phases; solid, liquid and gas (GLASS, 1989). The solid phase contains the major 

inorganic reserves of the soil, the liquid phase contains a source of nutrients that are 

immediately available for uptake by the roots and the gaseous phase permits 

exchange of gases, the most important of these gases being oxygen, carbon dioxide 

and nitrogen (GLASS, 1989). In addition to these physical phases soil also contains a 

wide variety of biota including a diverse community of interdependent plants, animals 

and microorganisms. Other definitions are that soils generally consists of the rocks 

and their weathering products, substances formed by reactions within the soil profile 

and material from plants and animals (SLEEMAN & BREWER, 1988) and that soil is 

a multilayer granular composite originating from larger rocks (LECLERC, 2003). Just 

by looking at the diversity in these explanations and definitions, it is clear that soil is 

very complex, and its study is very interdisciplinary. Although soil is so complex and 

creating a complete list of all the constituents of soil would be very difficult and has 

not been attempted, Figure 2.29 shows some of the general constituents of soil 

(SLEEMAN & BREWER, 1988) 

The mineral content of soil is the result of this weathering of larger rocks, climate, 

necrosis and decomposition of biota and the action of soil micro biota (LECLERC, 

2003). The organic matter content of soil originates from the necrosis and 

decomposition of cells, tissues, organs and whole organisms (LECLERC, 2003). The 

nature and type of a soil is determined by the rock it originates from, the climate and 

the surrounding biota (LECLERC, 2003). 

35


Figure 2.29: Diagrammatic representation of a small cluster of soil illustrating the complexity of 

organic soil. Also note the air spaces between the various components illustrated. Modified 

from descriptions of SLEEMAN & BREWER (1988). 

2.6.1 Physical properties of soil 

The physical properties of soils, which are greatly influenced by surface to volume 

relations of the soil particles, are important in determining the availability of various 

important ions to the plant roots (GLASS, 1989). Soil particles often have inorganic 

particles bound to their surface by a specific charge, making the ions unavailable to 

the plant for uptake (GLASS, 1989). When bulk flows are insufficient to supply the 

plant demand, diffusion-limited zones might develop around the roots. This results in 

fluctuations in the amounts of various mineral nutrients available to the plant 

(GLASS, 1989). 

The physical properties of soil include texture, structure, density, porosity, water 

content, consistency, temperature and colour (DONAHUE et al., 1983). These 

influence root penetration and the availability of water and oxygen to the root surface 

(DONAHUE et al., 1983). Only soil texture and water content will be discussed here. 

36

2.6.1.1 Soil texture 


Soil is generally comprised of soil particles of various sizes (DONAHUE et al., 1983). 

Soil texture can be described by granulometry, which is the quantification of the 

distribution of these soil particles in the different size classes (LECLERC, 2003). 

These size classes are called the soil separates and they can be placed in three 

general classes: Sands, silts and clays (Table 2.1) (DONAHUE et al., 1983). 

Table 2.1: Names of the soil separates and the particle diameters which define them (Modified 

from DONAHUE et al. (1983)). 

Soil separate name Diameter range (mm) 

Stones > 254 

Cobbles 75 to 254 

Gravels 2 to 75 

Very coarse sand 1.0 to 2.0 

Coarse sand 0.5 to 1.0 

Medium sand 0.25 to 0.5 

Fine sand 0.10 to 0.25 

Very fine sand 0.5 to 0.15 

Silt 0.002 to 0.5 

Clay < 0.002 

The texture of a soil can be determined after the percentage of each separate within 

a sample is known and these are then grouped into percentage sand, silt and clay 

(DONAHUE et al., 1983). These percentages can then be plotted on a triangular 

graph (See Figure 2.30). The description of the soil is determined by drawing three 

lines each perpendicular to a side of the triangle and arranged on the axes according 

to the relative percentages. The description at the point where these lines meet can 

then be read off. This method can also be used to determine the percentage content 

of a third separate group if the percentage content of 2 is known. Soil texture has a 

large influence on plant growth due to its effect on soil water retention capacity, 

oxygen capacity and thermal conductivity (LECLERC, 2003). Particles larger than 2 

mm but less than 250 mm also play a large role in soil texture (DONAHUE et al., 

1983). When classifying a soil, the names of these separates precede the rest of the 

name (e.g. ‘Stony’, ‘silty’ and ‘clay’). 

37


Figure 2.30: A textural triangle showing the range of variation in sand, silt, and clay for each 

soil textural class (Modified from DONAHUE et al. (1983) and LOVELAND & WHALLEY (1991)). 

2.6.1.2 Soil water content 

Water is a solvent for the soil solution and is essential for plant growth. This depends 

on the available surface area within the soil, which is determined by soil texture 

(LECLERC, 2003). In general, soil water content refers to the water that is 

evaporated by heating soil at 100 to 110 C until no further weight loss is observed 

(GARDNER et al., 1991). This measurement does however not include structural 

water, which usually requires heating between 400 and 800 C to evaporate. An 

example of such structural water is the water molecules which are incorporated with 

hydroxyl groups in clay lattice structures (GARDNER et al., 1991). In soil water 

38


content studies involving organic soils, some inaccuracy is probable. In such cases 

some of the observed decrease in weight may be caused by changes in the 

composition of the organic matter (GARDNER et al., 1991) 

Field capacity is defined as the soil moisture content where gravitational drainage 

ceases in a soil that was saturated with water (DONAHUE et al., 1983). Wilting point 

is defined as the specific soil moisture content at which the plant can no longer 

absorb water. The water available to the plant can thus be calculated by the 

difference between field capacity and wilting point (DONAHUE et al., 1983). 

2.6.2 Organic matter 

This includes living, dead and decomposed biotic matter (DONAHUE et al., 1983). 

Soil organic matter is a very important factor in soil fertility (DONAHUE et al., 1983). 

It is a reservoir of essential plant nutrients, including nitrogen and phosphorous. Soil 

organic matter also loosens up the soil to provide aeration (DONAHUE et al., 1983). 

2.6.3 Soil nutrients 

Sixteen chemical elements are known to be important to a plant's growth and 

survival. The sixteen chemical elements can be divided into two main groups, non- 

mineral and mineral, according to chemical structure. The non-mineral nutrients are 

hydrogen (H), oxygen (O), & carbon (C). The mineral nutrients are divided into two 

groups; macronutrients and micronutrients (See Table 2.2). The macronutrients are 

considered to be the nutrients essential to plant growth. 

Table 2.2: Classification of mineral elements into macro- and micronutrients (Modified from 

(MARSCHNER, 1995)). 

Classification Element 

Macronutrient N, P, S, K, Mg, Ca 

Micronutrient Fe, Mn, Zn, Cu, B, Mo, Cl, Ni, Na, Si, Co 

39


Macronutrients can be divided into two more groups; primary and secondary 

nutrients. The primary nutrients are nitrogen (N), phosphorus (P), and potassium (K). 

These major nutrients usually are lacking from the soil first because plants use large 

amounts of these elements. The secondary nutrients are calcium (Ca), magnesium 

(Mg), and sulphur (S). There are usually enough of these nutrients in the soil. 

Micronutrients are those elements essential for plant growth which are needed in 

only very small quantities. These are boron (B), copper (Cu), iron (Fe), chloride (Cl), 

manganese (Mn), molybdenum (Mo) and zinc (Zn) (MARSCHNER, 1995). 

2.6.4 pH 

The abbreviation for pH comes from the French term pouvoir hydrog ne or 

“hydrogen power” (DONAHUE et al., 1983). This is because the amount of hydrogen 

ions is the variable measured by instruments used to determine the pH of a solution. 

The pH of a soil depends on the original weathered rock content and the C/N ratio of 

the surrounding biota and decomposing organic matter (LECLERC, 2003). The 

occurrence of ions of Ca and Mg in a soil is often correlated with a increase in pH 

(pH of 7.5 to 8.5) (DONAHUE et al., 1983; LECLERC, 2003). This is also correlated 

with a decrease in the concentrations of K and Na (DONAHUE et al., 1983). A very 

acidic soil is usually a result of extensive leaching, a high proportion of sesquioxide 

and kaolinite, slow microbial activity and a low concentration of exchangeable basic 

cations (DONAHUE et al., 1983). 

The pH of the nutrient solution affects the solubility and availability of nutritional 

elements (DONAHUE et al., 1983). Most minerals are more soluble in acidic soils 

(DONAHUE et al., 1983). Soluble forms of aluminium and manganese are commonly 

found in soils with a high acidity (pH of 4 to 5) (DONAHUE et al., 1983). A high acidity 

also negatively affects many nitrogen fixing bacteria (DONAHUE et al., 1983). 

A highly basic soil can also affect plant growth negatively (DONAHUE et al., 1983). 

Examples of such soils include soils that have a high calcium content and soil which 

have not been leached (DONAHUE et al., 1983). These are common in low rainfall 

areas such as Namaqualand (DONAHUE et al., 1983) 

40


Because processes that causes the production of H + ions contributes to acidity, 

anything that reduces the activity of H + ions in the soil solution would thus neutralize 

soil acidity. Sources of such alkalinity include low rainfall, resulting in less leaching of 

non-acid cations, increased sodium levels and plant uptake of nutrients. An effective 

neutralizer of acidity is calcium. Each mole of CaCO3 neutralizes 2 moles soil acid. It 

is the carbonate in CaCO3 that acts to neutralize soil acidity. Other neutralizers 

include magnesium carbonate, calcium oxide, calcium hydroxide and wood ash. An 

acid soil with a high cation exchange capacity needs a greater amount of limestone 

than a low cation exchange capacity soil of the same pH, because of the much 

greater number of reserve H + ions held in the soil with the high cation exchange 

capacity (HEWITT & SMITH, 1974). 

2.6.5 Salinity 

A saline soil is defined as a nonalkali soil containing soluble salts in quantities that 

interfere with the growth of common crop plants (GAUCH, 1972). Soil is generally 

seen as saline if it contains more than 0.1% soluble salts (GAUCH, 1972). 

2.6.6 Cation and anion exchange capacity and surface charges 

The cation exchange capacity of soil can be defined as the sum of positive charges 

of the absorbed cations that a soil can absorb at a specific pH (MARSCHNER, 1995). 

The more clay and organic matter a soil contains, the higher its cation exchange 

capacity and the stronger the cations are held. 

2.6.7 Soils of Namaqualand 

The soils of Namaqualand are very diverse. The occurrence of a impenetrable 

hardpan layer in most plain or valley landscapes is however a common feature of 

this area (DESMET, 2007). The soils of Namaqualand can be broadly classified into 

3 groups according to DESMET (2007). 

These are the weakly structured grey, yellow or red medium sands of the Sandveld 

and Bushmanland, the shallow, undifferentiated and free-draining red and yellow, 

variably grained, sandy to loamy soils of the Kamiesberg and Richtersveld ranges 

and the red, base-rich, granite-derived colluvial soils rich in clay of the inland margin 

of the coastal plain below the Hardeveld range (DESMET, 2007). 

41


The soils of the Sandveld and Bushmanland are derived from aeolean reworking of 

marine or fluvial deposits, whereas the soils of the Kamiesberg and Richtersveld 

range is derived from in situ weathering of the underlying parent material (DESMET, 

2007). The sands of the inland margin of the coastal plain below the Hardeveld range 

has been reworked through time to produce a variety of sand types, ranging from 

white, calcareous sands to deep red, acidic sands (DESMET, 2007). 

2.6.8 Soils of Nieuwoudtville 

R. sabulosa is restricted to the clay and sandy soils of the Bokkeveld escarpment 

west of Nieuwoudtville in renosterveld (MANNING & GOLDBLATT, 2001). Many 

other species in this genus including R. monadelpha and R. flava also occur in the 

vicinity of Nieuwoudtville, making this an area of interest. 

Not only the plant life, but also the soils of the Nieuwoudtville region are diverse 

(BRAGG et al., 2005). In the west, along the Bokkeveld escarpment exposed Table 

Mountain Sandstone (TMS) is prominent, while tillite and shales covers the 

underlying TMS in the east (BRAGG et al., 2005). Tillite is defined as a sedimentary 

rock composed of indurated (rendered hard by heat, pressure or cementation) till, 

which is an unstratified and unsorted sediment carried or deposited directly by or 

under a glacier. Shale can be defined as finely stratified consolidated sediment 

mainly composed of clay-sized particles (SOIL CLASSIFICATION WORKING 

GROUP, 1991). 

A dolerite sill intrudes into tillite in the vicinity of Nieuwoudtville (BRAGG et al., 2005). 

Dolerite is an igneous rock that has risen from under the earth crust as magma and 

mostly solidified as intrusions such as dykes and sills before reaching the surface 

(SOIL CLASSIFICATION WORKING GROUP, 1991). In this case the sill is visible as 

a rocky ridge. On the moist east slopes of this ridge, the dolerite has weathered 

through time to form self-mulching clay soils (BRAGG et al., 2005). 

42

2.7 PROPAGATION OF ROMULEA SPECIES 


There are very few publications on the propagation of Romulea species. In a book by 

DENO (1993), it was reported that R. bulbocdium and R. luthicii geminated at 10°C, 

with no germination at 20°C. It also stated that 96% germination was obtained during 

winter in Tauranga, New Zealand with fresh R. hantamnensis seeds, whereas seeds 

which had been in dry storage for an unknown time showed no germination (DENO, 

1993). The only scientific paper that could be found on this topic is a study on the 

seed dispersal and germination of R. rosea by EDDY & SMITH (1975). Some 

preliminary tests done by EDDY & SMITH (1975) suggests that application of KNO3 

and prechilling for 5 days at 2°C does not increase the germination of this species. 

These experiments indicated an optimum temperature of 10 to 11°C. This was 

confirmed by larger experiments, which showed that its germination has an optimum 

in the range of 9.5 to 15°C and is inhibited by temperatures exceeding 16.5°C. They 

found that germination of this species is quite slow relative to other species occurring 

in the pastures which they studied (EDDY & SMITH, 1975). They also showed that 

seed collected in 1969 had a higher percentage germination and shorter time to 

germination than seed collected in 1968 (EDDY & SMITH, 1975). This and the 

experiments on R. hantamnensis seeds described by DENO (1993) suggest that 

fresh seeds of species of this genus should be used for germination. 

2.8 GERMINATION PHYSIOLOGY 

2.8.1 Seed structure 

An angiosperm seed is typically comprised of the embryo, which is the result of 

fertilization of the egg cell in the embryo sac by a male pollen tube nucleus; the 

endosperm, which arises from the fusion of two nuclei in the embryo sac with the 

other pollen tube nucleus; the perisperm, which is developed from the nucellus; and 

the protective testa or seed coat, which is formed from one or both the integuments 

around the ovule (HARTMANN & KESTER, 1965; BEWLEY & BLACK, 1994). 

43


The embryo is the diploid result of fertilization and is a minute autotrophic plant 

(EDMOND et al., 1964). The embryo principally consists of an embryonic axis and at 

least one cotyledon (BEWLEY & BLACK, 1994). The axis further consists of the 

embryonic root (radicle), the hypocotyl with attached cotyledons and the shoot apex 

with the first true leaves (plumules) attached (BEWLEY & BLACK, 1994). 

The nourishing tissue is generally either the endosperm or cotyledons (EDMOND et 

al., 1964). In the case of endospermic seeds, the endosperm, which is present in the 

mature seed, serves as a food storage organ (HARTMANN & KESTER, 1965). Here 

the testa and endosperm are the two layers covering the embryo (BEWLEY & 

BLACK, 1994). In non-endospermic seeds the cotyledons serve as the sole food 

storage organ (BEWLEY & BLACK, 1994). During development, the cotyledons 

absorb the food reserves from the endosperm. Here the embryo is enclosed by the 

testa and the endosperm is all but completely degraded in the mature seed 

(BEWLEY & BLACK, 1994). 

In some cases the testa exists in a rudimentary form only and the prominent and 

outermost structure is the pericarp or fruit coat derived from the ovary wall 

(HARTMANN & KESTER, 1965). In such cases the embryo is also encased in a fruit 

(BEWLEY & BLACK, 1994). The seed coverings provide mechanical protection to the 

embryo and make transportation and storage of seeds possible (HARTMANN & 

KESTER, 1965). 

Hairs or wings, which aid in seed dispersal, sometimes develop as a modification of 

the enclosing fruit coat (BEWLEY & BLACK, 1994). These are attached via the hilum 

(BEWLEY & BLACK, 1994). The hilum is a funicular scar on the seed or fruit coat 

that marks the point at which the seed was attached via the funiculus to the ovary 

tissue (LAWRENCE, 2000). In many cases a small hole, called the micropyle, can be 

seen at the end opposite to the end with the hilum (BEWLEY & BLACK, 1994). 

Seeds store various substances that are important for germination and early seedling 

growth. These primarily include carbohydrates, fats and oils, and proteins (MAYER, 

1977; BEWLEY & BLACK, 1994). Other important substances that are only stored in 

small amounts include alkaloids, lectins, proteinase inhibitors, phytin, and raffinose 

44


oligosaccharides (RAGHAVAN, 1976; BEWLEY & BLACK, 1994). Most seeds store 

their major food reserves within the embryo; usually the cotyledons (BEWLEY & 

BLACK, 1994). Some plants also have their seed storage reserves within extra- 

embryonic tissues. These extra-embryonic tissues used for storage include the 

endosperm (Gymnosperms) or the perisperm (Coffea arabica). Both embryonic and 

extra-embryonic tissues can also be used for storage; such as in maize (BEWLEY & 

BLACK, 1994). 

2.8.2 Seed germination 

Germination starts with the uptake of water by a seed and ends with the onset of 

elongation of the embryonic axis, usually the radicle (BEWLEY & BLACK, 1994). It 

also includes the steps of protein hydration, sub cellular changes, respiration, 

macromolecular syntheses and cell elongation (RAGHAVAN, 1976; BEWLEY & 

BLACK, 1994). The combined effect of these steps is to transform a dehydrated, 

dormant embryo into an embryo which grows actively and accumulates water 

(MAYER, 1977; BEWLEY & BLACK, 1994). A seed in which none of these processes 

have taken place is said to be quiescent. They characteristically have low moisture 

content (5-15%) and an extremely slow metabolic rate (BEWLEY & BLACK, 1994). 

Seeds are able to survive in this state for a number of years. Quiescent seeds 

require an environment of suitable temperature, hydration and the presence of 

oxygen in order to germinate (BEWLEY & BLACK, 1994). The major cellular 

processes involved in the initiating and facilitating of radicle emergence include 

respiration, RNA and protein synthesis and enzyme and organelle activity 

(RAGHAVAN, 1976; BEWLEY & BLACK, 1994). 

During imbibition various structural and physical changes occur (BEWLEY & BLACK, 

1994). The completion of imbibition requires a small amount of water (not more than 

three times the seed’s dry weight). For successful subsequent root and shoot growth 

a larger and more constant supply of water is essential (BEWLEY & BLACK, 1994). 

The uptake of water by the seed can be seen as triphasic (Figure 2.31) (BEWLEY & 

BLACK, 1994). Phase 1 involves imbibition, where a strong osmotic gradient results 

in the uptake of water from the soil (BEWLEY & BLACK, 1994). Different areas of the 

testa and different organs within the seed often absorb variable amounts of water. 

45


Phase 2 is seen as a lag phase. Here the matrix forces of the seed cells that caused 

the strong osmotic gradient in phase 1 are no longer active (BEWLEY & BLACK, 

1994). During this phase major metabolic events take place in preparation for radicle 

emergence (BEWLEY & BLACK, 1994). Only germinating seeds, and not dormant 

seeds, enter phase 3. This stage is associated with changes in the cells of the radicle 

and radicle elongation (JANN & AMEN, 1977; BEWLEY & BLACK, 1994). These 

changes are facilitated by the uptake of water (JANN & AMEN, 1977). This uptake of 

water is a result of the production of low-molecular-weight osmotically active 

substances (BEWLEY & BLACK, 1994). These substances are produced as a result 

of hydrolysis of stored reserves (BEWLEY & BLACK, 1994). The kinetics of water 

uptake is however more complex than this, as many seeds distribute water to 

different seed parts at different rates. 

Figure 2.31: The triphasic pattern of water uptake by germinating seeds, with arrow showing 

the time of radicle protrusion (BEWLEY & BLACK, 1994). 

For germination to be completed, the radicle must expand and penetrate the 

surrounding structures (BEWLEY & BLACK, 1994). This does not require cell 

division. Instead, the radicle cells elongate as the radicle penetrates through the 

surrounding tissues and cell division starts some time after the testa is eventually 

ruptured. There are a number of possible requirements for radicle elongation 

(BEWLEY & BLACK, 1994). One such requirement is the lowering of the osmotic 

potential as a result of the accumulation of solutes within the radicle; this increases 

water uptake and raises the turgor pressure, which facilitates cell elongation. The 

46


initial growth of the seedling follows one of two distinct patterns (HARTMANN & 

KESTER, 1965). The seedling either follows the pattern of epigeous germination, 

where the hypocotyl elongates and raises the cotyledons above the ground, or 

hypogeous germination, where the lengthening of the hypocotyl does not cause the 

cotyledons to rise above the ground and only the epicotyl emerges (HARTMANN & 

KESTER, 1965). 

2.8.3 Measuring germination 

It is incorrect to equate germination to seedling emergence from soil, as germination 

ends sometime before this (BEWLEY & BLACK, 1994). Emergence of the axis can 

however be used as a precise measurement of termination of germination (BEWLEY 

& BLACK, 1994). 

The progress of germination is expressed as a percentage of the total number of 

seeds tested at time intervals throughout the germination period (BEWLEY & 

BLACK, 1994). When this relationship is expressed graphically it ordinarily yields a 

sigmoid curve. Some valuable conclusions can be drawn from variations in the shape 

of such a curve. If the curve flattens off when only a low percentage of the seeds 

have germinated it indicates that the seeds have a low germinating capacity 

(BEWLEY & BLACK, 1994). The shape of the curve also describes the uniformity of 

germination (BEWLEY & BLACK, 1994). 

Mean germination time can be calculated by the following equation: MGT= (n×d) /N. 

Here n=number of seeds germinated on each day, d=number of days from the 

beginning of the test, and N=total number of seeds germinated at the termination of 

the experiment (ELLIS & ROBERTS, 1981). When this is combined with a measure 

of germination, data can be displayed in a more concise way (KULKARNI et al., 

2007). 

2.8.4 Promotion and inhibition of germination 

2.8.4.1 Gibberellin and abscisic acid 

Numerous studies have shown that GA promotes germination in dormant and non- 

dormant seeds (JONES & STODDART, 1977). It has also been shown that levels of 

47


endogenous gibberellins increase during low temperature treatments (JONES & 

STODDART, 1977). 

Gibberellins (GA) promote the induction of cell wall hydrolases and thereby promote 

endosperm weakening and endosperm rupture (MARION-POLL, 1997). Abscisic acid 

(ABA) inhibits the induction of cell wall hydrolases and thereby inhibits this 

weakening and rupture (MARION-POLL, 1997). 

GA promotes and ABA inhibits the embryo growth potential. ABA, however, also 

plays an important role in seed development and germination, acquisition of 

desiccation tolerance, accumulation of proteins and lipid reserves, and induction and 

maintenance of seed dormancy (MARION-POLL, 1997). 

The expression of genes encoding enzymes that mobilise food reserves is induced 

by several known GA signalling factors. These food reserves include the starches, 

proteins and lipids that are stored in the endosperm (PENG & HARBERD, 2002). 

During protein rehydration GA biosynthesis is induced by a phytochrome-mediated 

light signal and the newly synthesised GA down-regulates the expression of protein 

repressors of germination. These are also activated by rehydration, by protein 

degradation, suppression of transcription or mRNA degradation (MARION-POLL, 

1997). This newly synthesized GA also initiates signals to induce the expression of 

hydrolytic enzymes that modify the cell wall and weaken the endosperm cap, thus 

facilitating germination (PENG & HARBERD, 2002). 

It has also been proposed that GA could promote the formation of low molecular 

weight mono- and disaccharides, which assist the intracellular generation of negative 

water potentials, thus aiding radicle emergence (TIAN et al., 2003). 

2.8.4.3 Cytokinins and auxins 

Cytokinins are involved in cell division and thus both radicle and cotyledon expansion 

(EMERY & ATKINS, 2006). Cytokinins accumulates mainly in the endosperm of 

seeds (LEUBNER-METZGER, 2006). Auxins are involved in the coordination of 

correct cellular patterning after the globular stage in embryogenesis (LEUBNER- 

METZGER, 2006). 

48

2.8.4.2 Potassium nitrate 


This chemical is commonly used to encourage seed germination, with solutions of 

0.1 and 1.0% frequently being used in germination tests (COPELAND, 1976). KNO3 

has been shown to act synergistically with factors such as temperature and media 

supplementation with kinetin and gibberellic acid. Seeds sensitive to KNO3 are often 

also sensitive to light (COPELAND, 1976). 

2.8.4.3 Ethylene 

Increased ethylene evolution accompanies seed germination of many species 

(BASKIN & BASKIN, 1998). Ethylene promotes its own biosynthesis during pea seed 

germination by positive feedback regulation of 1-aminocyclopropane-1-carboxylic 

acid oxidase. 

2.8.4.4 Smoke 

Many plant species have shown increased percentage germination and vigour after 

exposure of seeds or seedlings to smoke water (MINORSKY, 2002). These include 

plants naturally adapted to areas of high fire frequency as well as some species that 

are not specifically adapted to these conditions. These include agriculturally useful 

species, such as maize (SPARG et al., 2006) and lettuce (BROWN & VAN STADEN, 

1997), as well as many aesthetically useful species, such as those naturally 

occurring in fynbos (BROWN & VAN STADEN, 1997). Species originating from 

natural fire-prone habitats include many species of South African fynbos such as the 

fire-climax grass, Themeda triandra (Poaceae) and members of the 

Mesembryanthemaceae (BROWN & VAN STADEN, 1997). Other examples of such 

smoke water stimulated plants include species of the California chaparral and many 

other fire-prone communities (BLANK & YOUNG, 1998). It has been suggested that 

the promotive effect of smoke is independent of seed size and shape, plant life form 

and fire sensitivity (BROWN & VAN STADEN, 1997). Smoke can also be utilised as 

an effective seed pre-sowing treatment, as the stimulatory effect of smoke is 

irreversible and cannot be leached (LIGHT et al., 2002). Seeds treated with smoke 

are known to retain this stimulatory effect even after a year of storage (MINORSKY, 

2002). The inhibitory effects of high smoke concentrations appear to be reversible 

and seeds grow with increased vigour after the smoke has been leached to a 

49


tolerable level (LIGHT et al., 2002). This effect would be a favourable adaptation to a 

post-fire environment as the inhibitory compounds will only leach with sufficient 

rainfall (LIGHT et al., 2002). Smoke thus enables seeds to germinate at the right 

time, grow faster and have a more robust root system and so have a major 

competitive advantage in their natural environment (BLANK & YOUNG, 1998). 

In early experiments by DE LANGE and BOUCHER (1990), smoke was generated by 

burning a mixture of fresh and dry plant material in a metal drum. This smoke was 

then fed into a polythene tent and allowed to settle on the soil where the seeds were 

stimulated to germinate. One disadvantage is that the germination cue of smoke is 

easily confused with the effect of temperature on germination. This is because 

temperatures slightly higher than ambient temperature significantly increase seedling 

emergence in many species (BASKIN & BASKIN, 1998). Due to the complications of 

separating smoke from high temperatures, direct exposure to smoke is not 

recommended. 

Aqueous smoke extracts were pioneered by DE LANGE and BOUCHER in 1990 and 

since then many authors have demonstrated that the active component of airborne 

smoke is soluble in water (BROWN & VAN STADEN, 1997; TAYLOR & VAN 

STADEN, 1998; SPARG et al., 2005). The method for preparing such a solution 

usually involves forcing smoke that has been generated in a drum to bubble through 

water (BROWN & VAN STADEN, 1997). Combustion usually proceeds slowly and 

the burning material is made to smoulder, thus releasing relatively large quantities of 

smoke (BROWN & VAN STADEN, 1997). 

In an experiment conducted by JÄGER et al. (1996) it was found that aqueous 

smoke extracts prepared from a range of plants, as well as extracts prepared by 

heating agar and cellulose contained compounds which stimulated the germination of 

Grand Rapids lettuce seed. They also demonstrated that the same active compound 

is produced by burning Themeda triandra leaves, agar and cellulose by providing 

evidence obtained by thin-layer chromatography and high-performance liquid 

chromatography. This was also demonstrated with the same methods by BROWN 

and VAN STADEN (1997). 

50


BLANK and YOUNG (1998) reported that smoke increases the permeability to 

solutes of a subdermal seed membrane for some species of California chaparral. 

They also stated that these specific mechanisms of fire cue stimulation may be 

species dependent. This response involves triggering via elevated nutrient content or 

via the presence of stimulating gases in the smoke or triggering chemicals that 

permeate the embryo and induce enzymatic changes that trigger germination. In an 

experiment done by BROWN and VAN STADEN (1997) the dormancy of celery 

seeds was broken by a combination of plant-derived smoke, benzyladenine and 

gibberellins in the dark at temperatures between 18 and 26°C. From these results it 

could be argued that smoke extracts act in a similar way to cytokinins in the celery 

seed as it enhances gibberellin activity. 

In a study conducted at the ultrastructure level by EGERTON-WARBURTON (1998), 

the causal factor(s) associated with seed dormancy and the stimulation of 

germination were investigated for Emmenanthe penduliflora (Hydrophyllaceae) 

seeds. It was found that a short exposure to smoke resulted in two major 

morphological changes. These changes are closely associated with the stimulation 

and acceleration of germination. The first major and most visible smoke induced 

morphological change observed was an intense chemical scarification of the external 

cuticle (EGERTON-WARBURTON, 1998). This has a direct and destabilising effect 

on the external cuticle and is manifested as the formation of oil-like spheres or 

micelles. These micelles increase the surface area of the seed for the exchange of 

water and solutes, as well as altering the hydrophobicity of the seed surface 

(EGERTON-WARBURTON, 1998). 

The second, more important smoke induced morphological change occurred at the 

internal cuticle. Here the exposure to smoke stimulus caused a significant increase in 

the number and diameter of permeate channels in the cuticle of Emmenanthe 

penduliflora seeds (EGERTON-WARBURTON, 1998). The creation of such channels 

within the cuticle increases the permeability of this layer. It was shown through the 

use of the fluorescent apoplastic tracer dye, lucifer yellow (carbohydrazide), that 

51


these channels permit the rapid exchange of water and solutes between the external 

environment and the seed. 

Even though these major morphological changes have a large influence on the 

function of the cell, they do not significantly alter the general shape or dimensions of 

the cells (EGERTON-WARBURTON, 1998). EGERTON-WARBURTON (1998) 

speculated that the mechanism by which such morphological changes benefit the 

plant may be a synergy between the observed increased cuticular permeability and a 

simultaneous leaching of endogenous inhibitors of germination during imbibition. 

They also further described the mechanism of formation of permeate channels. 

Because the permeate channels occurs in the cuticle, the formation of channels has 

to be the result of a preceding reaction between certain constituents of smoke and 

the semi-crystalline structure of waxes (EGERTON-WARBURTON, 1998). This is a 

rather promising argument as the pyrolysis of cellulose alone produces a collection of 

compounds, such as aromatic hydrocarbons, ketones and a number of organic acids, 

which all have the potential to dissolve or modify waxes. 

Smoke also contains a number of compounds that may act as surfactants (e.g. 

alcohols) (EGERTON-WARBURTON, 1998). These surfactants modify cuticular 

layers by plasticizing the molecular structure of waxes (EGERTON-WARBURTON, 

1998). The sorption of such surfactants to the cuticle also creates hydrophilic 

channels. This increases the area of channels within the cuticle that leads to 

accelerated transcuticular penetration of solutes in several species. 

Using liquid chromatography, BLANK and YOUNG (1998) discerned over 30 organic 

anions in aqueous extracts of soil heated between 250 to 450°C. BLANK and 

YOUNG (1998) speculated that these unknown compounds or combinations of 

compounds could be cueing agents. 

FLEMATTI et al. (2004) identified the UV absorbance maximum and molecular 

weight of a germination-enhancing compound in smoke. They also assigned a 

molecular formula of C8H6O3 to the compound. They also identified the compound as 

the butenolide, 3-methyl-2H-furo[2,3-c]pyran-2-one and were able to synthesize it. 

52


They compared the activity of this synthesized butenolide with that of smoke water 

dilutions and found a high similarity in the germination delivered by these two 

compounds. Since this study, many experiments have been conducted using 

butenolide. This compound has been shown to enhance germination in a similar way 

to smoke water dilutions in many genera including Acacia, Eucomis and Dioscorea 

(KULKARNI et al., 2006b; KULKARNI et al., 2006a; KULKARNI et al., 2007). In these 

experiments a 10 −7 M butenolide solution was used. 

2.8.5 Phytochromes and light quality 

Photo-sensitive substances that are responsible for photoperiodic control of 

germination, in some species, were first discovered in 1959 (COPELAND, 1976). 

There are two photo-reversible forms of phytochrome in plants (COPELAND, 1976). 

The first is PR phytochrome, which is sensitive to orange-red light (600-680 nm), and 

the second is PF-R phytochrome, which is sensitive to far-red light (700-760 nm). In 

most species the greatest promotion of germination occur after exposure to light in 

the red area (660 to 700 nm) with a peak of germination often being observed at 670 

nm (COPELAND, 1976). Wavelengths at 440 nm, above 700 nm and below 290 nm 

are known to inhibit germination. No studies has shown an effect of wavelengths of 

290 to 400 nm on germination (COPELAND, 1976) 

2.8.6 Scarification 

Mechanical scarification 

To mechanically scarify seeds a small hole is made in the seed coat using a needle 

or a scalpel or the whole seed coat is scarified with sandpaper or specialized 

equipment (BASKIN & BASKIN, 1998). 

Acid scarification 

This is usually done by soaking seeds in concentrated sulphuric acid for a short 

period of time after which they are rinsed several times with distilled water (BASKIN 

& BASKIN, 1998). By using acid scarification, both the testa and the seed pores are 

scarified (BASKIN & BASKIN, 1998). 

53


2.8.7 Seed dormancy and the influence of temperature and stratification 

The imposition of dormancy is normally controlled endogenously and germination is 

initiated in response to a certain combination of environmental variables. These 

environmental variables include temperature, availability of minerals, light and 

weathering (KOLLER, 1972). With the termination of dormancy, the metabolic 

processes of synthesis and growth are resumed (KOLLER, 1972). Some seeds 

exhibit no dormancy, such as mangrove seeds which germinate on the tree itself, 

while other seeds such as lotus or lupin seeds may remain dormant for centuries or 

millennia (THIMANN, 1977). 

There are, generally speaking, two types of organic seed dormancy; endogenous 

and exogenous (BASKIN & BASKIN, 1998). In endogenous dormancy there is some 

characteristic of the embryo that prevents germination, while in exogenous dormancy 

it is some characteristic of the structures that cover the embryo that prevents 

germination. These structures include the endosperm, perisperm, testa and fruit walls 

(BASKIN & BASKIN, 1998). Seeds may, for example, be unable to germinate 

because of seed or fruit coats that are impermeable to water (BASKIN & BASKIN, 

1998). Before water uptake and subsequent germination can take place these blocks 

to germination must be removed. There are a number of endogenous and 

exogenous dormancy types (Table 2.3). 

Endogenous physiological dormancy is generally caused by a physiologically 

inhibiting mechanism of the embryo that prevents germination. The structures that 

cover the embryo may however also play a substantial role (BASKIN & BASKIN, 

1998). Physiological dormancy can be differentiated into non-deep, intermediate and 

deep physiological dormancy (BASKIN & BASKIN, 1998). Embryos of seeds in non- 

deep and intermediate dormancy tend to germinate when isolated from the 

surrounding tissues while those of seeds in deep physiological dormancy do not 

(BASKIN & BASKIN, 1998). 

54


Table 2.3: Organic seed endogenous and exogenous dormancy types (Modified from BASKIN & 

BASKIN (1998)). 

Endogenous 

Exogenous 

Type Cause Broken by 

Physiological 

Physiological inhibiting mechanism 

(PIM) of germination 

Morphological Underdeveloped embryo 

Morphophysiological 

Physical 

PIM of germination and 

underdeveloped embryo 

Seed/fruit coats impermeable to 

water 

Warm/cold stratification 

Appropriate conditions for 

embryo germination/growth 

Warm/cold stratification 

Opening of specialized 

structures 

Chemical Germination inhibitors Leaching 

Mechanical Woody structures restrict growth Warm/cold stratification 

The causes of non-deep physiological dormancy are factors relating to the covering 

structure (BASKIN & BASKIN, 1998). These factors include the physical barrier 

created by these structures, the resulting oxygen supply to the embryo, inhibitors 

within the covering structures and changes in the covering structures (BASKIN & 

BASKIN, 1998). Iris lorteti is an example of a species with seeds exhibiting non-deep 

physiological dormancy as a result of the physical restriction caused by their seed 

coats (BASKIN & BASKIN, 1998). It takes a force of 133.2 MPa, which can only be 

overcome by an embryo with sufficient growth potential, to break the seed coat of this 

species (BASKIN & BASKIN, 1998). 

The dormancy of such seeds can often be broken by a cold or hot stratification 

treatment (COPELAND, 1976). Such a treatment is performed by placing moistened 

seeds at low temperatures (3 to 10°C) for a certain period of time (COPELAND, 

1976). In some cases (European ash seed) dormancy can only be overcome by 

stratification (COPELAND, 1976). In such cases the growth-stimulating substance 

produced during stratification breaks dormancy caused by inhibitory chemicals within 

55


the embryo (COPELAND, 1976). Stratification has also been known to decrease the 

time to germination and increase growth rate in other species (COPELAND, 1976). 

Stratification may also decrease the sensitivity to external conditions, so that a seed 

might, for example, germinate at a less suitable temperature (COPELAND, 1976). 

The germination rate may also be improved by exposing seeds to different day/night 

temperatures (COPELAND, 1976). In some cases embryonic dormancy may be 

broken by exposure to a certain light intensity, wavelength and/or photoperiod 

(COPELAND, 1976). In some cases dormancy caused by inhibitory chemicals within 

the embryo can be broken by gibberellic acid (COPELAND, 1976). Other chemicals 

which are known to break non-deep physiological dormancy include potassium 

nitrate, kinetin and ethylene. The effects of endogenous physical dormancy also 

diminishes as seed age increases (COPELAND, 1976). 

In the case of morphological dormancy, germination is prevented at the time of 

maturity due to the morphological characteristics of the embryo. The embryo is 

underdeveloped or even undifferentiated at the time of dispersal and a period of 

growth, known as after-ripening, is required before the seed can successfully 

germinate (COPELAND, 1976; BASKIN & BASKIN, 1998). Morphological dormancy 

occurs in seeds with rudimentary and linear embryos. Most of the interior of these 

seeds is occupied by endosperm and the embryo may only be 1% of the seed 

volume or less (BASKIN & BASKIN, 1998). 

Morphophysiological dormancy is essentially a combination of the two dormancy 

types. In this case the embryo must grow to a species-specific critical size and the 

physiological dormancy of the seed must occur before germination can take place 

(BASKIN & BASKIN, 1998). 

The primary reason for dormancy of seeds with exogenous physical dormancy is the 

impermeability of their seed or fruit coats to water (BASKIN & BASKIN, 1998). Seeds 

with physical dormancy frequently have a palisade layer of lignified cells in the testa 

or pericarp (BASKIN & BASKIN, 1998). The breakdown of such hard seed coats in 

the natural environment occurs through the gradual processes of hydration and 

dehydration, exposure to hot and cold temperatures, scorching by fire and 

56


degradation due to extreme soil acidity, microbial breakdown and digestion by 

animals (COPELAND, 1976). Another reason for exogenous physical dormancy is 

sometimes the impermeability of seed coats to gasses (KHAN, 1977). Physical 

dormancy is often found in combination with chemical dormancy. 

Chemically dormant seeds do not germinate due to the presence of inhibitors in the 

pericarp. These inhibitors are usually removed by leaching (BASKIN & BASKIN, 

1998). These inhibitors include chemicals such as ABA (Abscisic acid) and phenolic 

compounds (KHAN, 1977). Many studies have shown that there is a interaction 

between phytochromes activated by red light and inhibitors (KHAN, 1977). 

Mechanical dormancy is usually the result of a hard, woody fruit wall or seed coat. 

This woody structure is usually the endocarp or the mesocarp (BASKIN & BASKIN, 

1998). Upon germination the endocarp often splits into two halves (BASKIN & 

BASKIN, 1998). In such cases dormancy can sometimes be broken with a period of 

cold stratification (KHAN, 1977). 

2.8.8 Seed longevity and viability 

Some seeds are viable after several years, decades or even after a few hundred 

years (BEWLEY & BLACK, 1982). Longevity is largely dependent on storage 

conditions. Factors that influence the longevity of seeds in storage include 

temperature, moisture and oxygen pressure (BEWLEY & BLACK, 1982). A low 

temperature and moisture content usually equates to a longer period of viability 

(BEWLEY & BLACK, 1982). Higher oxygen pressure results in a shorter period of 

sustained viability (BEWLEY & BLACK, 1982). Unorthodox or recalcitrant seeds 

cannot withstand drying (BEWLEY & BLACK, 1982). These seeds must retain a 

relatively high moisture content to remain viable (BEWLEY & BLACK, 1982). Even in 

relatively moist storage conditions they are rarely viable for more than a few months 

(BEWLEY & BLACK, 1982). The majority of seed plants are however orthodox and 

can remain viable for a prolonged period under suitable storage conditions (BEWLEY 

& BLACK, 1982). Various mathematical equations have been derived to relate the 

viability of seeds with their storage environment (BEWLEY & BLACK, 1982). To test 

the longevity of seeds the relative number of normal seedlings produced by seed 

57


germinated under controlled conditions each year serves as a comparative measure 

of seed viability (HARTMANN & KESTER, 1965). 

The tetrazolium test for seed viability is used by soaking seeds in a solution of 2,3,5- 

triphenyltetrazolium chloride (TTC) (HARTMANN & KESTER, 1965). This chemical is 

absorbed by living tissue and changed into an insoluble red compound, formazan, by 

NADPH dehydrogenases (HARTMANN & KESTER, 1965; COPELAND, 1976; 

LEADEM, 1984; BAND & HENDRY, 1993). Non-living tissue remains uncoloured 

(HARTMANN & KESTER, 1965). The reaction takes place equally well in dormant 

and non-dormant seeds and results are obtained in less than 24 hours. The test is 

used as a rapid assessment of viability or as a viability test of dormant seeds that do 

not respond to other methods (HARTMANN & KESTER, 1965; INTERNATIONAL 

SEED TESTING ASSOCIATION, 1999a). A 1% solution is commonly used, although 

a 0.05% solution may sometimes be satisfactory (HARTMANN & KESTER, 1965; 

BAND & HENDRY, 1993). It should be used at a pH of 6.5 to 7.5 (INTERNATIONAL 

SEED TESTING ASSOCIATION, 1999a). 

The intact seeds of some species may be soaked in a TTC solution whereas some 

seeds require to be soaked in water first so that tissues are hydrated (COPELAND, 

1976). Some seeds should be soaked in a solution with a respiration stimulant such 

as hydrogen peroxide (COPELAND, 1976). Other seeds require procedures to be 

followed before staining that include the removal of any hard covering and sectioning 

of the seeds so that the embryo may be exposed to the TTC solution (HARTMANN & 

KESTER, 1965). The embryo is then incubated in 1% TTC for 2 hours in the dark 

after which the excess tetrazolium is washed off with water (COPELAND, 1976; 

BAND & HENDRY, 1993). 

The amount of staining is then observed. The location and intensity of the formazan 

stain is important to accurately define the viability of the embryo (COPELAND, 1976; 

LEADEM, 1984). If the areas of cell division are unstained or abnormally stained the 

potential for germination to occur is lowered (COPELAND, 1976). A number of broad 

classes, defining the germinability of the embryo are given in Table 2.4. 

58

Table 2.4: Topographic stain evaluation classes for the TTC test (LEADEM, 1984). 

Class Description Viability 

1 Embryo completely stained Germinable 

2 Very pale staining Possibly germinable 


3 Cotyledons unstained Non-germinable or possibly germinable 

4 Radicle unstained Non-germinable or probably not germinable 

5 No staining Non-germinable 

The meristems should be well stained in order to insure healthy germination and 

growth of the embryo (LEADEM, 1984). A typical viable embryo should be at least 

75% stained and its tissue should be firm with a smooth surface (LEADEM, 1984). 

The TTC stain test leaves a number of uncertainties. A sample may, for example, not 

be stained because the stain never penetrates the tissue and the non-enzymatic 

reduction of TTC is also possible in dead and living tissue (BAND & HENDRY, 1993). 

A second viability test should be used to confirm the tetrazolium result (BAND & 

HENDRY, 1993). 

2.8.9 After-ripening 

In the period of after-ripening individual or collective changes take place so that a 

seed that was once dormant can germinate (COPELAND, 1976). This period may be 

determined by physical and chemical stimuli in the environment such as a 

phytochrome response and or the balance of inhibiting and promoting substances in 

the environment (COPELAND, 1976). It may also be determined by the effect of the 

environment on the balance of inhibiting and promoting substances within the seed 

and embryo itself and morphological growth and development of the embryo 

(COPELAND, 1976). 

After-ripening can sometimes be hastened by growth promoting substances, low- 

temperature stratification, alternating temperature and light exposure treatments 

59


(COPELAND, 1976). Such growth promoting substances are synthesised by the 

seed during this after-ripening process (KHAN, 1977). These include gibberellins, 

cytokinins and auxins. 

2.8.10 Embryo-excision as a tool for investigating mechanisms behind 

dormancy and testing viability 

Embryo excision serves as a good test for determining whether seed dormancy is 

endogenous or exogenous. This test can also be used to test seeds in cases where 

embryos require long periods of after-ripening before germination will take place 

(HARTMANN & KESTER, 1965). It is also useful to determine the viability of slow 

germinating seed (INTERNATIONAL SEED TESTING ASSOCIATION, 1999b). 

The embryo is essentially excised from the seed and germinated. Before excision is 

attempted, seeds must be soaked thoroughly, changing the water once or twice daily 

to avoid the accumulation of seed exudates, to retard the growth of contaminants and 

to circumvent the evolution of anoxic conditions (INTERNATIONAL SEED TESTING 

ASSOCIATION, 1999b). 

During the excision the embryo should be kept moist and working conditions should 

be aseptic. Observations such as predated, empty, decayed and discoloured seeds 

and deformed embryos should be noted and included in a calculation of viability 

(INTERNATIONAL SEED TESTING ASSOCIATION, 1999b). 

A viable embryo shows some indication of germination, whereas a non-viable 

embryo becomes discoloured and deteriorates (INTERNATIONAL SEED TESTING 

ASSOCIATION, 1999b). Signs of germination include the expansion of cotyledons, 

the development of chlorophyll and the growth of the radicle and plumules. The time 

required for this test ranges from 3 days to 3 weeks (HARTMANN & KESTER, 1965). 

Viability is calculated by dividing the number of viable embryos with the total number 

of seed tested and is reported as a percentage (INTERNATIONAL SEED TESTING 

ASSOCIATION, 1999b). 

60

2.8.11 Germination, dormancy and germination ecology in Iridaceae 


Geophytes are a very important component of Eurasian semi-deserts with cold 

winters. Geophytes belonging to Iridaceae in this biome exhibits morphophysiological 

dormancy (BASKIN & BASKIN, 1998). Such seeds are known to have 

underdeveloped embryos (BASKIN & BASKIN, 1998). Halophytes and emergent 

aquatics in the family of Iridaceae also exhibits morphophysiological dormancy 

(BASKIN & BASKIN, 1998). Iris angustifolia, Iris pseudacorus, Iris versicolor and Iris 

virginica are examples of such species (BASKIN & BASKIN, 1998). Two genera of 

the Iridaceae have been found in persistent seed banks (BASKIN & BASKIN, 1998). 

Accordingly, a species of Iridaceae could have morphological and/or 

morphophysiological dormancy. In a study by DIXON et al. (1995) it was found that a 

species in Iridaceae, Patersonia occidental, only germinated when treated with 

smoke. 

2.8.12 Embryo and seedling morphology of Iridaceae 

Embryo’s of the Iridaceae are straight and poorly differentiated (TILLICH, 2003). Only 

at the seedling stage can three different groups, defined by their cotyledon 

morphology, be clearly distinguished (TILLICH, 2003). The three groups are the 

compact cotyledon group, the tubular cotyledon group and the assimilating cotyledon 

group (TILLICH, 2003). 

At the event of germination the cotyledonary sheath, hypocotyl and radicle are 

pushed through the micropyle region of the testa (TILLICH, 2003). The cotyledon 

then immediately bends at 90° in most cases (TILLICH, 2003). In Crocoideae the 

typical cotyledon is characterised by a remarkable elongation of the cotyledonary 

sheath and / or the development of a long coleoptile (TILLICH, 2003). 

The seedling morphology of Crocus and Romulea is quite rare (TILLICH, 2003). Here 

an elongated tubular structure is combined with a tubular cataphyll (TILLICH, 2003). 

Possible explanations for this is the pronounced hypogeous germination which is 

promoted by a strong contractile primary root (TILLICH, 2003). The cotyledons of 

these genera have also lost most of their ability to produce chlorophyll and appear 

white even in high light intensities (TILLICH, 2003). 

61

2.9 BRIEF REVIEW OF IN VITRO CULTURE 


The term ”tissue culture” is actually a misnomer inherited from the field of animal 

tissue culture. Plant micropropagation involves the culture of a whole individual from 

isolated tissues, while animal tissue culture involves the culture of isolated tissues 

(KYTE & KLEYN, 1996). 

According to AHLOOWALIA et al. (2002), the process of micropropagation can be 

divided into five stages: the pre-propagation step (stage 0); the initiation of explants 

(stage I); the subculture of explants for multiplication or proliferation (stage II); 

shooting and rooting of the explants (stage III); and hardening off the cultured 

individuals (stage IV). The pre-propagation stage involves preparing the explant for 

aseptic culture. 

The explant and its response in vitro is significantly influenced by the phytosanitary 

and physiological conditions of the donor plant (KANE, 2004). Plant material used in 

clonal propagation should be taken from mother plants that have undergone the 

appropriate pre-treatment with fungicides and pesticides to minimize contamination in 

the in vitro cultures (AHLOOWALIA & PRAKASH, 2002). Such a piece of plant 

material is called an explant (SMITH, 2000b). A single explant can theoretically 

produce an infinite number of plants (KYTE & KLEYN, 1996). Explants can be 

obtained from meristems, shoot tips, macerated stem pieces, nodes, buds, flowers, 

peduncle pieces, anthers, petals, pieces of leaf or petiole, seeds, nucellus tissue, 

embryos, seedlings, hypocotyls, bulblets, bulb scales, cormels, radicles, stolons, 

rhizome tips, root pieces or protoplasts (KYTE & KLEYN, 1996). The explants should 

be surface decontaminated with antibiotic sprays before they are introduced into 

culture (AHLOOWALIA et al., 2002; KANE, 2004). 

In stage I an aseptic culture is initiated by inoculating the explant onto a sterile 

medium (AHLOOWALIA et al., 2002). Once such an explant is established it can be 

multiplied a number of times (AHLOOWALIA et al., 2002). The explants are then 

transferred to a contaminant free in vitro environment (AHLOOWALIA et al., 2002). 

62


In stage II or the propagation phase explants are cultured onto a medium that 

promotes the multiplication of shoots (AHLOOWALIA et al., 2002). Propagation must 

be achieved without excessive mutation (AHLOOWALIA et al., 2002). The culture of 

various organs in stage I lead to the multiplication of propagules in large numbers. 

These propagules can be cultured further and used for multiplication (AHLOOWALIA 

et al., 2002). These cultured shoots are often placed onto different media for 

elongation (AHLOOWALIA et al., 2002). 

The result of stage III is the production of complete plants, as the shoots derived from 

stage II are rooted (AHLOOWALIA et al., 2002). If shoot clumps are present, they 

should be separated after rooting. Many plants can be rooted on half strength 

Murashige and Skoog medium without any growth regulators (AHLOOWALIA et al., 

2002). Successful and sufficient rooting is essential for survival of the plant during 

hardening and transfer to the soil (AHLOOWALIA et al., 2002). 

The complete plants are weaned and hardened during stage IV (AHLOOWALIA et 

al., 2002). The plants should at this stage be autotrophic. Hardening consists of 

gradually altering the humidity, light and nutrition available to the plant. The plant is 

moved gradually from a high to a low humidity, from a low light intensity to a high 

light intensity and the agar is removed by gently washing it away with water. After 

sufficient hardening, plants can be transplanted to a suitable substrate and hardened 

further (AHLOOWALIA et al., 2002). 

In vitro regeneration techniques are essential to the application of in vitro selection 

techniques (REMOTTI & LÖFFLER, 1995). This is not only because it enables the 

selected genotype to be regenerated, but also it aids commercialisation of new 

species and selected genotypes (DEBERGH, 1994). 

Micropropagation enables the production of disease free plantlets at high rates and 

generally increases the efficiency of known breeding techniques (DEBERGH, 1994). 

It is also essential in the breeding of plants for which no breeding methods or 

procedures have been established (DEBERGH, 1994). In vitro selection techniques 

have been used to increase genetic variability and to broaden the gene pool 

63


(DEBERGH, 1994). In vitro techniques such as embryo rescue, protoplast fusion and 

genetic transformation enable plant breeders to accomplish wider crosses 

(DEBERGH, 1994). 

2.9.1 Explant selection 

There are numerous variables that should be considered when selecting an explant 

for in vitro culture. These are mostly caused by the physiology of the plant and its 

environment. 

Physiologically younger tissues are generally much more responsive to tissue culture 

(SMITH, 2000b). In many cases, older tissues will not form callus that is capable of 

regeneration. Younger tissue is usually the newest formed and therefore easier to 

surface disinfect (KYTE & KLEYN, 1996; SMITH, 2000b). Plant material at the base 

of a plant may, however be more suitable than explants higher up (KYTE & KLEYN, 

1996). The smaller the explant, the harder it is to culture (SMITH, 2000b). Larger 

explants have more nutrient and plant growth regulator reserves to sustain the 

culture. A large explant is often more difficult to decontaminate (KYTE & KLEYN, 

1996; SMITH, 2000b). Explants should be obtained from plants that are healthy as 

opposed to plants under nutritional or water stress or plants exhibiting disease 

symptoms (SMITH, 2000b). Plant material in a state of active growth is cleaner, and 

more suitable for aseptic culture, compared to dormant tissue. To control 

contamination, donor plants should be pre-screened for diseases (SMITH, 2000b; 

AHLOOWALIA et al., 2002). 

The season of the year can have an effect on contamination and the response of the 

explant in culture (SMITH, 2000b). Contamination tends to increase as summer 

progresses. Plant material obtained from the field is often more contaminated than 

material obtained from greenhouses or growth chambers (KYTE & KLEYN, 1996; 

SMITH, 2000a). Mother plants should ideally be maintained under dust, insect and 

disease free conditions (AHLOOWALIA et al., 2002). These plants should also not 

be stressed and they should preferably be grown under controlled conditions that 

promotes active growth (KANE, 2004). Such conditions should preferably include 

conditions of low relative humidity. Drip irrigation should preferably be used as the 

64


misting will facilitate contamination by wetting the foliage (KANE, 2004). Placing the 

plant material in a less humid and dry environment a few weeks prior to taking 

explant material can reduce contamination of cultures (SMITH, 2000b). 

Plant materials growing in soil (roots, tubers, bulbs) or near the soil surface (stolons, 

rhizomes, orchid protocorms, etc.) are often harder to clean and disinfect than aerial 

plant material (SMITH, 2000b). Explants are much easier to clean if the plant has 

been growing in an artificial medium, such as washed sand or perlite (KYTE & 

KLEYN, 1996). After cutting the explant from the source plant, it should be placed in 

a plastic bag containing a moist paper towel and kept refrigerated until culture 

initiation (KYTE & KLEYN, 1996). 

2.9.2 Explant preparation 

Explants require surface-disinfection before they can be placed in culture on the 

nutrient agar for in vitro culture (SMITH, 2000b). Explants are washed in sterile water 

and rinsed in ethanol and the surface is sterilised using chemicals with a chlorine 

base (AHLOOWALIA et al., 2002). There are a number of products used for surface 

disinfection, the most commonly used is commercial chlorine bleach (SMITH, 2000b). 

For soft, herbaceous material a calcium or sodium hypochlorite based solution is 

often used at a concentration of 1-3% (AHLOOWALIA et al., 2002). Before surface- 

disinfection any remaining soil or dead parts should be removed from the explant 

(PIERIK, 1997). An inexpensive and ready-made alternative is a 5-7% solution of 

Domestos® (a toilet disinfectant by Lever Bros. Ltd., UK), which contains 10.5% 

sodium hypochlorite, 0.3% sodium carbonate, 10.0% sodium chloride and 0.5% 

sodium hydroxide and a patented thickener (AHLOOWALIA et al., 2002). Explants 

are washed in sterile water before and after sterilization (AHLOOWALIA et al., 2002). 

A general procedure for preparing the explant involves washing the explant in warm, 

soapy water after which it is rinsed in tap water (PIERIK, 1997; SMITH, 2000b). The 

explant is then rinsed in a freshly made chlorine bleach solution (PIERIK, 1997). One 

to 2 drops of wetting agent should be added to every 100 ml of bleach solution. The 

explant is then rinsed in sterile water three to five times. 

65


PIERIK (1997) stated that a brief alcohol rinse or swab is needed with hairy or wax 

coated surfaces. Epidermal hairs may trap air bubbles, in such cases these have to 

be evacuated under vacuum. 

Sterilized forceps and scalpels must be used for the transfer of explants to fresh 

solutions (AHLOOWALIA et al., 2002). Sterile containers must be used throughout 

the protocol of surface sterilization (AHLOOWALIA et al., 2002). If explants become 

brown or pale the strength of the sterilizing agent should be reduced (GAMBORG & 

PHILLIPS, 1995; PIERIK, 1997). 

A cut explant such as a stem or leaf that is surface sterilised often shows tissue 

damage from surface sterilisation (PIERIK, 1997). The damaged tissue should be 

removed before culture (SMITH, 2000b). 

GAMBORG & PHILLIPS (1995) suggest that a procedure for seed sterilization should 

include washing the seeds in detergent, after which they are rinsed with tap water 

and subsequently alcohol, a bleach solution and autoclaved demineralised water 

respectively. 

Seeds can be germinated on filter paper in Petri dishes or on an agar medium 

(GAMBORG & PHILLIPS, 1995). A single seed should ideally be placed in each 

container so that a single contaminated seed does not contaminate other seeds. 

Contamination resulting from improperly sterilised tissue will generally arise from the 

explant and be located in the medium adjacent to the explant (SMITH, 2000b). 

Contamination that is due to poor technique will generally appear over the entire agar 

surface (SMITH, 2000b). Examples of poor technique include contaminated transfer 

hood filters and culture cabinets and improperly sterilised media. 

Contamination of cultures by fungi appear as fuzzy growth whereas bacterial 

contamination appears as smooth pink, white or yellow colonies and contamination 

66


from insects appears as tracks across the medium which are visible due to 

surrounding fungal or bacterial growth (SMITH, 2000b). 

Explant material may harbour internal micro-organisms (SMITH, 2000b). In such a 

case, it is very difficult to establish clean cultures. Explants that are least likely to 

harbour internal contaminants include explants taken from growing shoot tips, ovules 

of immature fruit, immature and mature flower parts and runner tips (SMITH, 2000b). 

Explants that are more likely to harbour internal contaminants include explants taken 

from bulbs, slow-growing shoots or dormant buds, roots, corms and underground 

rhizomes (PIERIK, 1997; SMITH, 2000b). In such cases seeds are often aseptically 

germinated to provide clean explants from the root, hypocotyl, cotyledon and shoot 

(PIERIK, 1997). In these situations the use of antibiotics or fungicides in the medium 

is generally not useful (SMITH, 2000b). Although these agents can repress the 

growth of some microorganisms, they can also suppress the growth of the plant 

tissue or even kill it (SMITH, 2000b). 

2.9.3 Medium composition 

A number of standard formulae for tissue culture media have been developed to 

provide optimum nutrients and growth regulators for specific plants (KYTE & KLEYN, 

1996). The selection or development of a suitable culture medium is vital to the 

success of the culture (SMITH, 2000b). The approach to the development of a 

suitable medium will depend on the purpose of the culture (SMITH, 2000b). 

The medium generally contains water, inorganic salts, plant growth regulators, 

vitamins, a carbohydrate and a gelling agent (SMITH, 2000b). High quality water 

should be used as an ingredient of the plant culture media (PIERIK, 1997; BEYL, 

2005). Ordinary tap water contains cations, anions, particulates, micro-organisms 

and gases that may influence the reaction of the tissue culture media with the tissue 

(BEYL, 2005). The most commonly used method of water purification involves a 

deionization treatment followed by one or two glass distillations (PIERIK, 1997; 

BEYL, 2005). 

67


The distinguishing feature of MURASHIGE & SKOOG (1962) inorganic salts is their 

high content of nitrate, potassium and ammonium in comparison to other salt 

formulations (SMITH, 2000b). Table 2.5 shows the composition of the Murashige and 

Skoog formula. MURASHIGE & SKOOG (MS) (1962) is the most suitable and the 

most commonly used basic tissue culture medium for plant regeneration from tissues 

and callus (BEYL, 2005). 

Table 2.5: The standard MURASHIGE & SKOOG (1962) formula. 

MAJOR SALTS 

MINOR SALTS 

IRON 

mg/l 

Ammonium nitrate 4 3 NO NH 1650 

Calcium chloride 

Magnesium sulphate 

CaCl2 2 2 

MgSO4 7 2 

⋅ H O 440 

⋅ H O 370 

Potassium nitrate KNO 3 1900 

Potassium phosphate KH 2PO 4 170 

Subtotal 4530 

Boric acid 3 3 BO H 6.2 

Cobalt chloride 

Cupric sulphate 

Manganese sulphate 

CoCl2 6 2 

CuSO4 5 2 

MnSO 2 

⋅ H O 0.025 

⋅ H O 0.025 

4 ⋅ H O 16.9 

Potassium iodide KI 0.83 

Sodium molybdate 

Zinc sulphate 

Na2 2 

MoO4 

⋅ 2H 

O 0.25 

ZnSO4 7 2 

⋅ H O 8.6 

Subtotal 32.83 

Ferrous sulphate 

FeSO4 7 2 

⋅ H O 27.8 

Na2EDTA 37.3 

Subtotal 65.1 

Total mg/l 4627.93 

68


Plant growth and developmental processes are controlled by plant growth regulators 

(GABA, 2004). The study of plant growth regulator function is complex because 

several plant growth regulators usually work in concert with each other and their 

concentration within plant tissues changes with time, season and developmental 

stage (GABA, 2004). The effect of plant growth regulators on plant growth and 

development depend on the chemical structure of the plant growth regulators used, 

the plant tissue used and the genotype of the plant (GABA, 2004). The type and 

concentration of the plant growth regulators used will vary according to the culture 

purpose (PIERIK, 1997). 

2.9.3.1 Auxins 

Auxins (for example: IAA, NAA, 2,4-D, or IBA) is required by most plants for cell 

division and root initiation (PIERIK, 1997; SMITH, 2000b). IAA or indole-3-acetic acid, 

was the first plant growth regulator to be isolated (GABA, 2004). IAA is rapidly 

degraded in growth media and inside the plant (GABA, 2004). For this reason 

chemical analogues of IAA with similar biological activity are often substituted 

(GABA, 2004). These more stable synthetic auxins include 2,4-D, IBA and NAA 

(GABA, 2004). IAA is added at a concentration of 0.01 to 10 mg l -1 , while synthetic 

auxins, such as IBA, NAA and 2,4-D, are used at concentrations of 0.001 to 10 mg l -1 

(PIERIK, 1997). IAA can be considered to be a weak auxin (PIERIK, 1997). Cultures 

in which a large quantity of IAA has been added are often less successful than 

cultures to which low concentration of a stronger auxin, such as NAA have been 

added (PIERIK, 1997). 

At high concentrations auxin can suppress morphogenesis (SMITH, 2000b). Auxins 

have numerous effects on plant growth and differentiation, depending on their 

chemical structure, their concentration and the affected plant tissue (GABA, 2004). 

Auxins generally stimulate cell elongation, cell division in cambium tissue and, 

together with cytokinins, the differentiation of phloem and xylem and the formation of 

adventitious roots (PIERIK, 1997). High concentrations of auxins can induce somatic 

embryogenesis (GABA, 2004). 

69


The essential function of auxins and cytokinins is to reprogram somatic cells that 

were in a state of differentiation (GABA, 2004). Such reprogramming causes 

dedifferentiation and then redifferentiation into a new developmental pathway (GABA, 

2004).The mechanism of dedifferentiation is not understood (GABA, 2004). 

The use of 2,4-D should be avoided, as it may induce mutations (PIERIK, 1997). This 

growth regulator is however important in callus initiation for many species. 

2,4-D is widely used for callus induction, while IAA (0.6-60 µM), IBA (2.5-15 µM) and 

NAA (0.25-6 µM) are mainly used in root initiation (SMITH, 2000b; GABA, 2004). 

Higher than optimum levels of auxins causes callus production and a reduction in 

root growth and root quality (GABA, 2004). Combinations of auxins at low 

concentrations can sometimes produce better results than using individual auxins 

(GABA, 2004). High concentrations of auxins are sometimes necessary to induce 

rooting (GABA, 2004). This can however have undesirable side effects such as 

growth inhibition of induced roots (GABA, 2004). In such cases the elevated auxin 

levels should be administered as a pulse treatment (GABA, 2004). To do this the 

shoot is incubated with auxin for several days before it is transferred to a medium 

with no plant growth regulators to allow root growth and development (GABA, 2004). 

Somatic embryogenesis is typically induced by auxins, sometimes in combination 

with cytokinins (GABA, 2004). 2,4-D is commonly used at this stage, although other 

auxins can be used (GABA, 2004). Auxins induce the cells to become embryogenic 

and promote subsequent repetitive cell division of embryogenic cells (GABA, 2004). 

High concentrations of auxins prevent subsequent cell differentiation and embryo 

growth (PIERIK, 1997). 

2.9.3.2 Cytokinins 

As the name suggests, cytokinins cause cell division (GABA, 2004). Such cell 

division can lead to shoot regeneration in vitro, by stimulating the formation of shoot 

apical meristems and shoot buds (PIERIK, 1997). This cell division caused by 

cytokinins can also cause the production of undifferentiated callus (GABA, 2004). A 

high concentration of cytokinins can cause the release of shoot apical dominance 

70


and will block root development (PIERIK, 1997). Examples of cytokinins include 

kinetin, zeatin, 2-iP, BA and thidiazuron (PIERIK, 1997). Some new cytokinins, 

called topolins have been synthesised in the Laboratory of Growth Regulators, 

Palacký University and Institute of Experimental Botany AS CR, Czech Republic. 

These include MemT [6-(3-methoxybenzylamino)purine, MemTR [6-(3- 

methoxybenzylamino)-9-b-D-ribofuranosylpurine], mT [6-(3- 

hydroxybenzylamino)purine]and mTR [6-(3-hydroxybenzylamino)-9-b-D- 

ribofuranosylpurine] (BAIRU et al., 2007). 

A high concentration of cytokinins can cause many small shoots to initiate but fail to 

develop (GABA, 2004). Shoots are induced into forming roots by placing them in a 

regeneration medium, containing a high level of cytokinin, and then in a medium with 

no plant growth regulators (GABA, 2004). Cytokinins inhibit rooting and can be 

effectively removed from the plant material by placing shoots in a medium without 

plant growth regulators (GABA, 2004). Such a treatment can also be used to reduce 

endogenous cytokinin levels (GABA, 2004). 

2.9.3.3 Gibberellins 

Gibberellins are, in most cases, non-essential for plant development in in vitro culture 

(PIERIK, 1997). In tissue culture, gibberellic acid (GA3) is used to stimulate either 

shoot elongation or the conversion of buds into shoots (PIERIK, 1997). Gibberellins 

reduce root formation and embryogenesis in vitro (PIERIK, 1997). Gibberellins are 

primarily used to stimulate cell elongation and to produce elongated shoots in plant 

tissue culture (GABA, 2004). Unwanted side effects caused by gibberellins include 

reduction in the number of buds produced, the elongation of leaf structures such as 

petioles and lamina, the excessive elongation of shoots and reduced root production 

(GABA, 2004). 

YASMIN et al. (2003) found that GA3 dissolved in 0.2% ethanol inhibited adventitious 

rooting of mungbean cuttings but when dissolved in water, GA3 promoted 

adventitious rooting at 10 -7 M and 10 -8 M (YASMIN et al., 2003). This indicates that 

ethanol suppresses the promoting effects of GA3 (YASMIN et al., 2003). 

71

2.9.3.4 Abscisic acid 


Abscisic acid is rarely used in tissue culture protocols and has a negative effect on 

growth in most cases (PIERIK, 1997). It is mainly used in plant tissue culture to 

facilitate somatic embryo maturation (GABA, 2004) but may also be used in some 

regeneration processes and rarely used to produce somatic embryos (GABA, 2004). 

ABA induces the formation of essential LEA (late embryogenesis abundant) proteins 

found at late stages of embryogenesis in somatic and sexual embryos (GABA, 2004). 

LEA proteins are associated with tolerance to water stress resulting from desiccation 

and cold shock (GOYAL et al., 2005). 

2.9.3.5 Ethylene 

Ethylene or physiological reactions similar to that caused by ethylene, is produced by 

certain plastic containers, plant tissue and as a result of fire (PIERIK, 1997). This is 

the only gaseous natural plant growth regulator and it is naturally produced by all 

plant tissues in a controlled fashion (GABA, 2004). Endogenously produced ethylene 

can accumulate in a closed vessel to levels that negatively affect plant growth and 

development (PIERIK, 1997). The biological effect of ethylene depends on how air- 

tight the vessel is and the sensitivity of the plant material (GABA, 2004). 

Ethylene is primarily known for its effects on fruit ripening (GABA, 2004). Exposure to 

ethylene also results in reduced stem length, restricted leaf growth, premature leaf 

senescence and may cause increased growth of axillary buds (GABA, 2004). An 

enhanced ethylene concentration can induce callus formation, while inhibiting bud 

and shoot regeneration (GABA, 2004). Low concentrations of ethylene stimulate 

somatic embryogenesis, while high concentrations of ethylene inhibit somatic 

embryogenesis (GABA, 2004). Explants need a low level of ethylene for correct 

biological functioning, but too high an ethylene concentration leads to symptoms of 

excess (GABA, 2004). 

Such symptoms of excess include stunted growth, a reduction in leaf size and leaf 

drop (PIERIK, 1997; NOWAK & PRUSKI, 2002). These plants do not acclimatise well 

to the in vivo environment and often desiccate shortly after being transferred to soil 

(NOWAK & PRUSKI, 2002). 

72


Endogenous ethylene has an important role in shoot and root growth and 

differentiation (PIERIK, 1997). 

2.9.3.6 Combinations of plant regulators 

A high ratio of auxin to cytokinin induces root formation in shoots of dicotyledonous 

plants and somatic embryogenesis (GABA, 2004). Intermediate ratios induce callus 

initiation and adventitious root formation from callus in dicotyledonous plants (GABA, 

2004). Such intermediate ratios often involve high levels of both cytokinins and 

auxins (GABA, 2004). Low ratios of auxin to cytokinin induce adventitious shoot 

formation and axillary shoot production in shoot cultures (GABA, 2004). 

The optimum cytokinin to auxin ratio can be established by using a matrix approach 

(Table 2.6) with two axes of increasing plant growth substance concentration (KYTE 

& KLEYN, 1996). 

73

Table 2.6: Example of a matrix to establish optimal auxin to cytokinin ratios and their 


concentrations, where the rows represent auxin levels and the columns represent the cytokinin 

levels (Modified from Kyte and Kleyn (1996). 

0 

0.5 

1 

3 

5 

10 

2.9.3.7 Vitamins 

0 0.5 1 3 5 10 

The vitamin considered most important for plant cells is thiamine (B1) (SMITH, 

2000b). Other vitamins, such as nicotinic acid (B3) and pyridoxine (B6), are also 

added to culture media, as they may enhance cellular response (SMITH, 2000b). 

2.9.3.8 Carbohydrates 

Green cells in culture are generally not photosynthetically active and require a carbon 

source (SMITH, 2000b). Sucrose or glucose at 2-5% (w/v) is commonly used 

(SMITH, 2000b). Higher levels of sucrose leads to low levels of photosynthesis in the 

leaves (ROBERTS et al., 1990). Higher levels may however be used for embryo 

culture (SMITH, 2000b). Sugars undergo caramelisation when autoclaved too long 

(SMITH, 2000b). When sugars are heated they degrade and form melanoidins, which 

are brown, high molecular weight compounds that can inhibit cell growth (SMITH, 

2000b). 

2.9.3.9 Gelling agent 

The type of agar used to gel the medium can affect the response of experiments 

(SMITH, 2000b). To minimise problems that arise from agar impurities, washed or 

purified agar should be used (SMITH, 2000b). 

74

2.9.4 Liquid culture 


In liquid culture, the explant is covered by the medium; enlarging the surface area for 

absorbing nutrients and plant growth regulators (ASCOUGH & FENNELL, 2004). The 

medium can be changed automatically, this reduces labour costs of subculturing 

(ASCOUGH & FENNELL, 2004). 

Liquid culture does however have a few side-effects (ASCOUGH & FENNELL, 2004). 

Because of the submersion of tissue, the explant may become oxygen deficient. This 

leads to the formation of elongated and hyperhydric leaves (ASCOUGH & FENNELL, 

2004). There are a few methods to overcome hyperhydricity, they are however not 

universal and some of them retard growth and multiplication (ZIV, 1989; ASCOUGH 

& FENNELL, 2004). 

Gladiolus bud explants have been propagated in an agitated liquid medium (ZIV, 

1989). ZIV (1989) concluded that liquid cultures can be used to scale up the 

micropropagation of Gladiolus sp. and possibly other geophytes as it allows for a 

faster rate of cormlet production (ZIV, 1989). 

2.9.5 Embryo-excision 

Embryo rescue is known as one of the earliest and most widely used techniques for 

Iridaceae micropropagation (KRIKORIAN & KANN, 1986). Embryo culture is a well 

established branch of in vitro culture and is known as one of the oldest and most 

successful culture procedures (HU & ZANETTINI, 1995; REED, 2005). In embryo 

rescue, the artificial medium substitutes for the endosperm (REED, 2005). 

MURASHIGE & SKOOG (1962) is the most frequently used basal media for embryo 

culture. 

Embryo development occurs in two phases, a heterotrophic and an autotrophic 

phase (REED, 2005). In the heterotrophic phase, the young embryo, or “proembryo”, 

requires a complex medium. In vivo grown embryos at this stage are dependent on 

the endosperm. Amino acids such as glutamine and asparagine are often added to 

the culture medium. 

75


Young embryos require a medium with high osmotic potential (PIERIK, 1997). A high 

osmotic potential prevents precocious development and promotes normal 

embryogenic development (REED, 2005). Sucrose is usually added to serve both as 

an osmoticum and a carbon source (PIERIK, 1997). A medium with 8 to 12% sucrose 

is used for the culture of heterotrophic embryos (REED, 2005). 

The autotrophic phase is usually initiated in the late heart-shaped embryo stage 

(REED, 2005). Embryos that are excised during this development stage are 

completely autotrophic (HU & ZANETTINI, 1995). Such embryos germinate and grow 

on a simple inorganic medium with a supplemental energy source (HU & ZANETTINI, 

1995). An inorganic medium supplemented with 2 to 3% sucrose is used as a 

standard medium for the germination of autotrophic embryos (REED, 2005). 

Growth regulators often have inconsistent effects on embryo culture (REED, 2005). 

They have however been extensively used in embryo rescue protocols, especially 

protocols involving heterotrophic embryos (REED, 2005). Low concentrations of 

auxins promote normal growth whereas gibberellins cause embryo enlargement and 

cytokinins inhibit growth (REED, 2005). 

Hard-coated seeds are first soaked in water for a few hours up to a few days before 

dissection. Seeds are surface sterilised before and after soaking (HU & ZANETTINI, 

1995). The most suitable point of incision into the ovule differs amongst species 

(REED, 2005). The embryos of some species can be extracted by cutting off the 

micropylar end of the ovule and applying gentle pressure at the opposite end of the 

ovule, so that the embryo is pushed through the opening (REED, 2005). Small seeds 

are dissected by making a longitudinal section using sterile microdissection needles 

(HU & ZANETTINI, 1995) 

After excision, large embryos should immediately be transferred into culture vessels, 

using a pair of forceps (HU & ZANETTINI, 1995; REED, 2005). Small embryos can 

be handled using the moistened tip of a dissection needle (HU & ZANETTINI, 1995). 

76

2.9.6 Callus culture 


Explants, when cultured in the appropriate medium, usually with both auxin and 

cytokinin, give rise to a mitotically active, but unorganized mass of cells. It is thought 

that, under the right conditions, any plant tissue can be used as an explant (SLATER 

et al., 2003). Callus culture concerns the initiation and continued proliferation of 

undifferentiated parenchyma cells from explant tissue on clearly defined semi-solid 

media (BROWN, 1990). 

Callus initiation is the first step in many tissue culture experiments (BROWN, 1990; 

SMITH, 2000b). In vivo, callus is a wound tissue produced in response to injury or 

infestation (BROWN, 1990; MINEO, 1990; SMITH, 2000b). Not all the cells in an 

explant contribute to callus formation (SMITH, 2000b). Only certain callus types, 

which are competent to regenerate organised structures, display totipotency (SMITH, 

2000b). 

The level of plant growth regulators is a major factor that controls callus formation in 

the culture medium (BROWN, 1990; SMITH, 2000b). The correct concentration of 

plant growth regulators depends on the species, individual and explant source 

(SMITH, 2000b). Other culture conditions such as light, temperature and media 

composition are also important for callus formation and development (SMITH, 

2000b). 

Explants can be taken from various plant organs, structures and tissues (MINEO, 

1990). Young tissues of one or a few cell types are most often used as explants 

(MINEO, 1990). The pith cells of a young stem are regarded as a good source of 

explant material for callus initiation (MINEO, 1990). 

Callus growth is maintained, provided that the callus is subcultured onto a fresh 

medium periodically. During callus formation there is a degree of de-differentiation in 

both morphology (usually unspecialised parenchyma cells) and metabolism. As a 

consequence most plant cultures lose their ability to photosynthesise (SLATER et al., 

2003). This means that the metabolic profile does not match that of the donor plant 

77


and the addition of compounds such as vitamins and a carbon source is necessary 

(SLATER et al., 2003). 

Callus culture is often performed in the dark as light can result in differentiation of the 

callus (SLATER et al., 2003). The culture often loses its requirement for auxin and/or 

cytokinin during long-term culture (SLATER et al., 2003). This process is known as 

habituation and is common in callus cultures from some species such as sugar beet 

(SLATER et al., 2003). 

By manipulating the auxin to cytokinin ratio whole plants can subsequently be 

produced from callus cultures (PHILLIPS et al., 1995). Callus culture can also be 

used to initiate cell-suspension cultures (PHILLIPS et al., 1995). 

Endogenous levels of plant growth regulators and polar growth regulator transport 

can drastically influence callus induction (PIERIK, 1997; SMITH, 2000b). Explant 

orientation and different sectioning methods affect callus induction (SMITH, 2000b). 

Callus cultures subcultured regularly on agar media exhibit a sigmoidal growth curve 

(PHILLIPS et al., 1995). PHILLIPS et al. (1995) describe five phases of callus growth. 

Phase I is a lag phase, where cells prepare to divide. Phase II is an exponential 

phase, where the rate of cell division is the highest. Phase III is a linear phase, where 

cell division slows, but the rate of cell expansion increases. Phase IV is a 

deceleration phase, where the rates of both cell division and expansion decrease and 

phase V is a stationary phase, where the number and size of cells remain constant. 

Callus growth can be monitored in a non-destructive manner using fresh weight 

measurements (STEPHAN-SARKISSIAN, 1990; PHILLIPS et al., 1995). Dry weight 

measurements are more accurate, but involve the destruction of the sample 

(PHILLIPS et al., 1995). Mitotic index measurements of cell division rates are not 

easy to perform as they require numerous measurements to be made at various time 

intervals with very small amounts of tissue (PHILLIPS et al., 1995). Fresh weight 

measurements are performed by culturing a known weight of callus for a given time 

78


(typically 4 weeks) and weighing the callus after this time using aseptic techniques 

(STEPHAN-SARKISSIAN, 1990). 

2.9.7 Organogenesis 

Organogenesis involves the de novo production of organs directly from an explant or 

through initial callus culture (SCHWARTZ et al., 2004). In the Iridaceae 

organogenesis involves the regeneration of unipolar meristems (ZIV, 1997). 

Organogenesis is regulated by altering the components of the culture medium 

(BROWN & CHARLWOOD, 1990). Most important of these components is the auxin 

to cytokinin ratio, which determines the developmental pathway the regenerating 

tissue will take (BROWN & CHARLWOOD, 1990). Shoots are usually induced to form 

first by increasing the cytokinin to auxin ratio of the culture medium (BROWN & 

CHARLWOOD, 1990). These shoots can then be easily rooted (SLATER et al., 

2003). 

De novo organ formation via indirect organogenesis, which involves intermediate 

callus formation and a differentiation phase, may increase the possibility for 

somaclonal variation (SCHWARTZ et al., 2004). Any stage in the process of 

organogenesis that involves callus growth should be minimized. 

After dedifferentiation the explant acquires a state of competence, defined as its 

ability to respond to organogenic stimuli (SCHWARTZ et al., 2004). The attainment of 

competence can not always be achieved with a single step. The induction phase 

occurs between the time of competence and determination (SCHWARTZ et al., 

2004). During induction, processes resulting from the expression of genes guides 

developmental processes and precede morphological differentiation. It has been 

suggested that such a genetically determined developmental process can be 

interrupted by certain physical and chemical stimuli (PIERIK, 1997). At the end of the 

induction phase, the cells are fully committed to the production of shoots or roots. At 

this point the tissue can be removed from the root or shoot producing medium and 

placed on a basal medium without plant growth regulators (PGR’s), containing 

mineral salts, vitamins and a carbon source (SCHWARTZ et al., 2004). The desired 

79


organ is then produced on this medium. Successful determination is partially 

dependent on the chemical and physical environment to which they have been 

exposed. The result of failure of explant tissues to express totipotency is a failure of 

the explant tissues to achieve the state of competence for induction. This makes 

investigation of the effects of physical and chemical parameters difficult. The use of 

biochemical or genetic markers that can clearly indicate the developmental 

disposition of the primary explant tissue have not yet been discovered (SCHWARTZ 

et al., 2004). 

In the next phase the morphological differentiation and development of the nascent 

organ occurs (SCHWARTZ et al., 2004). Organ initiation involves a rapid shift in 

polarity followed by a smoothing of this shift into a radially symmetrical organization 

and the concurrent growth along the new axis to form a characteristic bulge 

(SCHWARTZ et al., 2004). There is not an absolute certainty as to which tissues are 

involved and the number of cells involved in meristem initiation (SCHWARTZ et al., 

2004). 

The initiation of adventitious roots occurs in four stages (SCHWARTZ et al., 2004). A 

meristematic locus is first formed by the dedifferentiation of a stem or other cells. 

These cells then multiply to form a spherical cluster. Further cell multiplication occurs 

with the initiation of planar divisions to form a recognizable bilateral root meristem. 

Lastly, the cells located in the basal part of the developing meristem elongate, 

resulting in the eventual emergence of the newly formed root (SCHWARTZ et al., 

2004). The production of a functional adventitious root system depends on the 

selection of a microcutting of the appropriate developmental stage and the ability of 

the in vitro environment to initiate the sequence of events described above 

(SCHWARTZ et al., 2004). 

2.9.8 Somatic embryogenesis 

In plants, morphologically and functionally correct nonzygotic embryos can arise from 

an array of cell and tissue types at a number of different points within both the 

gametophytic and sporophytic phases of the plant life cycle (GRAY, 2005). Somatic 

(or asexual) embryogenesis involves the formation of embryo-like structures, which 

80


have the potential to develop into whole plants in a way analogous to zygote 

embryos, from somatic tissues (SLATER et al., 2003). Plant regeneration by somatic 

embryogenesis was first observed in carrot in 1958 (PHILLIPS et al., 1995). These 

somatic embryos can be produced either directly or indirectly. In direct somatic 

embryogenesis, the embryo is formed directly from a cell or a small group of cells 

without the production of an intervening callus (FINER, 1995). Direct somatic 

embryogenesis is however rare and only common in reproductive tissues (SLATER 

et al., 2003). In indirect somatic embryogenesis, a callus is first produced from the 

explant before embryos are produced. 

Synthetic auxins, especially 2,4-D, are most often used in protocols involving somatic 

embryogenesis (GRAY, 2005). Somatic embryogenesis usually proceeds in two 

distinct stages. In the initial stage (embryo initiation), a high concentration of 2,4-D is 

used (FINER, 1995). In the second stage (embryo production) embryos are produced 

in a medium with no or little 2,4-D (SLATER et al., 2003). 

Auxins activate pathways that induce the formation of embryogenic cells (GRAY, 

2005). Auxins promote division, while suppressing differentiation and growth of 

embryogenic cells (GRAY, 2005). When using embryogenic cells as an explant, 

auxins are often not required as there is no need for an induction step (GRAY, 2005). 

In many dicotyledonous species cytokinins are also required to induce 

embryogenesis (GRAY, 2005). BA is the cytokinin most often used in embryogenesis 

(GRAY, 2005). Other cytokinins that have been used include TDZ, kinetin and zeatin 

(GRAY, 2005). Somatic and microspore embryogenesis has been reported in Tulipa 

sp., Gladiolus sp. and Nerine sp. (ZIV, 1997). 

2.9.9 Hardening 

Micropropagation of a species is in some cases restricted due to unsuccessful ex 

vitro acclimatization, leading to a low survival rate of cultured plants 

(HUYLENBROECK et al., 2000). During this period of ex vitro acclimatization, plants 

must acquire the morphological and physiological features required by the in vivo 

environment and develop new patterns of resource allocation (HUYLENBROECK et 

al., 2000). In vivo cultured plants will otherwise not be able to cope with the 

81


environmental stresses of the post-propagation environment (HUYLENBROECK et 

al., 2000). 

During acclimatization there is a switch to autotrophy and changes in stomatal 

functioning and cuticular composition (HUYLENBROECK et al., 2000). Water is 

rapidly lost from the in vitro cultured plantlet because of the failure of the stomata to 

respond to stimuli that would normally induce their closure (ROBERTS et al., 1990). 

The poorly developed cuticle results in a rapid loss of water (ROBERTS et al., 1990). 

Vitrified plants do not acclimatise well to in vivo conditions. Vitrified plants are 

common where liquid media and low agar concentrations are used (PIERIK, 1997). 

In vitrified plantlets there is a reduced deposition of cellulose and lignin, leading to an 

increase in water uptake by the cells and resulting in glassy swollen leaves and 

stems (ROBERTS et al., 1990). Because of this, and the low rates of photosynthesis 

sustained by the in vitro cultured plants, they easily suffer from photoinhibition and 

water stress; leading to the production of reactive oxygen species 

(HUYLENBROECK et al., 2000). It has been demonstrated that micropropagated 

plants develop antioxidant mechanisms during acclimatization (HUYLENBROECK et 

al., 1998). 

In vitro grown leaves are the only source of nutrition to cover metabolic demands and 

to sustain plant adaptation and regrowth during the first days after transplanting 

micropropagated plants to greenhouse conditions (HUYLENBROECK et al., 1998). 

The good and sustainable health of leaves is therefore essential to the 

acclimatization and survival of the plant (HUYLENBROECK et al., 1998). 

Plant hardening is usually carried out under greenhouse conditions to increase the 

chance of survival (AHLOOWALIA & PRAKASH, 2002). A commonly used 

greenhouse is the Quonset type. This consists of movable or fixed benches with 

hardening tunnels on them (AHLOOWALIA & PRAKASH, 2002). It is also 

advantageous to acclimatise plants to lower humidities while they are still under in 

vitro conditions (AHLOOWALIA et al., 2002). In this way, plants grown in strongly 

82


aerated vessels often require little or no hardening (AHLOOWALIA & PRAKASH, 

2002). 

There are numerous differences between leaves formed in vitro and ex vitro: 

differences in wax composition, pigmentation content, stomatal response and 

photosynthetic performance (HUYLENBROECK et al., 1998). In a study conducted 

by HUYLENBROECK et al. (2000) on Calathea plants (Marantaceae) it was shown 

that chlorophyll and carotenoid content in leaves formed ex vitro were almost three 

times higher than in vitro. 

Stomatal aperture can be measured by applying nail varnish to the abaxial surface of 

the mature leaf (ROBERTS et al., 1990). The image of the hardened film can then be 

analysed. This allows us to assess the degree to which the plant has hardened to in 

vivo conditions (ROBERTS et al., 1990). 

2.9.10 Applications of in vitro culture 

In vitro culture and recombinant DNA technology enable plant improvement through 

sexual and para-sexual methods (AHLOOWALIA, 1997). Such methods include 

mutation induction, embryo rescue, anther and ovary culture, protoplast fusion and 

transgenic methods. Using these methods other flower colours, flower shapes and 

growth habits of species in the genus Romulea could possibly be obtained 

(AHLOOWALIA, 1997). 

2.10 CORM PHYSIOLOGY 

One main difference between corms and bulbs is that a corm is a stem base swollen 

with food reserves (HUSSEY, 1977a; KRIKORIAN & KANN, 1986). These stem 

bases have several nodes, each of which has its own superficial axillary bud 

(HUSSEY, 1977a). The leaves around a corm are distinctly different from those of 

bulbs. These leaves, which sheath the new corm, are thin and rarely useful as 

explants for culture initiation (HUSSEY, 1977a; KRIKORIAN & KANN, 1986). 

83

2.11 IN VITRO FLOWERING 


Florogenesis is divided into six phases for most geophytes. These include induction, 

initiation, differentiation, floral stalk elongation, maturation and growth of the floral 

organs, and anthesis. (FLAISHMAN & KAMENETSKY, 2006). Factors that affect 

florogenesis include the genetics of the plant and its environment (FLAISHMAN & 

KAMENETSKY, 2006). After induction of flowering the vegetative meristem ceases 

leaf production during flower initiation to allow for more resources for flower initiation 

(FLAISHMAN & KAMENETSKY, 2006). 

Smoke and smoke solutions have been used to promote flowering in geophytes. 

These include Narcissus tazetta, a freesia hybrid and Watsonia spp. (IMANISHI, 

1983; UYEMURA & IMANISHI, 1984; LIGHT et al., 2007). LIGHT et al., (2007) found 

that a 1:500 (v/v) smoke water solution promoted the flowering of corms of Watsonia 

spp., which is in the same subfamily as Romulea. Corms of Freesia sp., a species in 

the same tribe as Romulea, were stimulated to flower using a smoke treatment 

(UYEMURA & IMANISHI, 1984). 

In a study of the plant-soil relationship in the habitat of R. columnae it was found that 

soil N, P and K was reduced during the generative growth stage whereas the soil 

organic matter, pH, and CaCO3 levels were increased (KÖK et al., 2007). 

It is better to use large plants with large, matured storage organs for florogenesis 

studies, as it is common for a plant with a small storage organ not to flower 

(FLAISHMAN & KAMENETSKY, 2006). 

84

2.12 IN VITRO PROPAGATION OF GEOPHYTES 


ZIV (1997) defines geophytes as plants pereniating by underground storage organs. 

Geophytes are generally capable of developing tubers, corms or bulbs in vitro 

(STEINITZ & LILIEN-KIPNIS, 1989). Many ornamental geophytes are used for 

gardening, pot plant production, flowering pot plant production, cut flower production 

and the production of phytochemicals (ZIV, 1997). 

Plant biotechnology has provided geophyte production with clonal propagation, virus 

elimination, breeding and crop improvement through embryo rescue, in vitro 

fertilization, somaclonal variation, protoplast isolation and somatic hybridisation, and 

haploid production (ZIV, 1997). Genetic transformation is also aiding in geophyte 

development. Benefits include improving horticultural traits and induction of disease 

resistance (ZIV, 1997). Gene mapping and DNA fingerprinting are also additional 

developing areas (ZIV, 1997). 

Micropropagation has been achieved by enhanced axillary bud development, 

organogenesis and adventitious bud formation or by somatic embryogenesis (ZIV, 

1997). Explants used for propagation of bulbous, cormous and tuberous plants 

include the leaf lamina, petiole, mesophyll and epidermis (ZIV, 1997). Inflorescence 

peduncle, pedicel, tepals, petals, sepals, ovaries, anthers, ovules and embryos have 

also been used (ZIV, 1997). Other explants include the basal plate, scales, twin- 

scales and nodal and storage tissue of bulbs, corms and tubers (ZIV, 1997). 

The use of the underground pereniating organ as an explant source is often 

associated with heavy pathogen contamination (ZIV, 1997). This is a destructive use 

of the pereniating organ and eliminates the possibility of further vegetative or 

horticultural evaluation (ZIV, 1997). Different parts of the young flower and 

inflorescence stem can be cultured (ZIV, 1997). This is a source of pathogen-free 

totipotent explants (ZIV, 1997). Totipotency depends on the developmental stage at 

time of excision and position from which the explant was isolated (ZIV, 1997). 

Explants isolated from tissue positioned immediately next to the basal plate have a 

85


higher regenerative potential in some species of geophytes than explant obtained 

from tissue in the upper sections of the inflorescence stem (ZIV, 1997). 

2.13 IN VITRO PROPAGATION OF BULBOUS PLANTS 

Micropropagation protocols are available for many bulbous plants. The only bulbous 

species for which a large scale micropropagation program is however functional is 

Lilium sp (DEBERGH, 1994). Producing cormlets has an advantage over producing 

plantlets, as it prevents the development of vitreous leaves (ZIV, 1989). 

2.14 IN VITRO PROPAGATION OF IRIDACEOUS SPECIES 

Within one family there are often similar requirements or difficulties in 

micropropagation (KYTE & KLEYN, 1996). Understanding the culture requirements 

of species in the same family (Iridaceae) could shed some light on the culture 

conditions needed for the successful propagation of species in genus Romulea. An 

excellent review of Iridaceae micropropagation protocols has recently been published 

by ASCOUGH et al. (2009), this section will therefore only be considering studies 

involving direct shoot and corm organogenesis. These were chosen because they 

are the micropropagation steps used most commonly in this study and for species of 

Iridaceae. Direct shoot organogenesis and corm induction is summarized separately 

in Table 2.7 and 2.8 and is discussed separately in relation to results for some 

Romulea species in chapter five and six. 

86


Table 2.7: Explant sources and PGR's used by various authors for direct shoot or meristimoid 

organogenesis in genera of Iridaceae. Where the concentrations of PGR's are not mentioned, 

the study included multiple species within the genus, each reacted differently to various 

concentrations. A question mark indicates that the specific parameter is not included in the 

described publication. The genera are grouped phylogenetically, with vertical text on the right 

showing classification. 

Subfamily 

Ixioideae 

Tribe 

Ixieae 

Subtribe 

Romuleinae 

Babianinae 

Ixiinae 

Genus 

Crocus 

Babiana 

Ixia 

Dierama 

Sparaxis 

Author(s) 

BHAGYALAKSHMI (2000) Ovary 21.6 M NAA and 

22.2 M BA 

HOMES, LEGROS and JAZIRI 

(1987) 

PLESSNER, ZIV and NEGBI 

(1990) 

MCALISTER, JÄGER and VAN 

STADEN (1998) 

JÄGER, MCALISTER and VAN 

STADEN (1995) 

MEYER and VAN STADEN 

(1988) 

Explant (s) 

PGR(s) 

Corm 0.5 M kinetin and 

4.5 M 2,4-D; 

followed by 

subculture in 9.1 

M 2,4-D 

Corm 1.4 µM GA3, 2.3 

M 2,4-D and 2.3 

M kinetin 

Hypocotyl 4.4 M BA and 5.3 

M NAA 


M NAA 

Corm, Leaf 5 or 10 M BA 

SUTTER (1986) Corm 4.4 M BA 

KOETLE, FINNIE and VAN 

STADEN (2010) 

Seedling 

hypocotyl 

1.0 M zeatin 

MADUBANYA (2004) Hypocotyl 2.2 µM BA 

PAGE and VAN STADEN 

(1985) 

HUSSEY (1975) Corm, 

Inflorescence 

Corm 2.7 µM NAA / No 

PGR's 

0.2 to 0.7 µM NAA 

HUSSEY (1976) Shoot 0.5 µM BA 

87

Subfamily 

Ixioideae (continued) 

Tribe 

Ixieae (continued) 

Subtribe 

Gladiolinae 

Freesiinae 

Genus 

Gladiolus 

Freesia 

Author(s) 

DANTU and BHOJWANI (1987) Corm 2.2 µM BA 

DANTU and BHOJWANI (1995) Corm 2.2 µM BA 

DE BRUYN and FERREIRA 

(1992) 

Explant (s) 


Inflorescence 


PGR(s) 

Corm 2.2 µM to 8.9 µM 

BA 

0.2 to 0.7 µM NAA 


HUSSEY (1977a) Corm 0.5 to 1.3 µM BA 

JÄGER, MCALISTER and VAN 

STADEN (1998) 

KUMAR, SOOD, PALNI and 

GUPTA (1999) 

LILIEN-KIPNIS and KOCHBA 

(1987) 


M NAA 

Corm, 

Inflorescence 

2.5 M BA and 

10.0 M NAA 

Corm kinetin/BA and 

NAA 

TAN NHUT, TEIXEIRA DA 

SILVA, HUYEN and PAEK 

(2004) 

Corm 2.2 M BA 

SEN and SEN (1995). Corm 1.0 mg.l-1 BA 

STEINITZ, COHEN, 

GOLDBERG and KOCHBA 

(1991) 

STEINTZ and LILIEN-KIPNIS 

(1989) 

Corm 0.3 M BA and 

0.1 M NAA 

Corm 0.3 M BA and 

0.1 M NAA 

SUTTER (1986) Corm 4.4 µM BA 

ZIV and LILIEN-KIPNIS (2000) Inflorescence 10 M kinetin and 

5 M NAA 

ZIV (1970) Corm 2.7 M kinetin and 

2.3 M NAA 

ZIV (1989) Corm 10 M BA and 

0.25 M NAA 

ZIV, RONEN and RAVIV (1998) Corm 0.5 M NAA, 0.5- 

5.0 M BA and 1.7 

M PP3 


Inflorescence 

0.2 to 0.7 µM NAA 

HUSSEY (1976) Shoot 2.2 to 35.5 µM BA 

HUSSEY (1977b) Corm 2.2 to 8.9 µM BA 

88

Subfamily 


Iridoideae 

Tribe 


Watsonieae 

Sisiyrinchieae 

Irideae 

Tigrideae 

Hesperanthinae 

Subtribe 

Tritoniinae 

(Unknown) 

(Unknown) 

Iridinae 

Cipurinae 

Shizostylis 

Genus 

Tritonia 

Crocosmia 

Watsonia 

Sisyrinchium 

Iris 

Cipura 

Author(s) 

Explant (s) 


HUSSEY (1975) Inflorescence 0.2 to 0.7 µM NAA 


ASCOUGH, SWART, FINNIE 

and VAN STADEN (2011) 

GEORGE and SHERRINGTON 

(1984) 

ASCOUGH, ERWIN, VAN 

STADEN (2007) 

ASCOUGH, SWART, FINNIE 

and VAN STADEN (2011) 

Seedling 3.3 µM mT 

? ? 

Hypocotyl BA and NAA 

PGR(s) 

Seedling 4.1 to 20.7 µM mT 


YABUYA, YOSHIHARA, INOUE 

and SHIMIZU (2006) 

Shoot 4.4 µM BA and 5.4 

µM NAA 

HUSSEY (1977b) Corm 2.2 to 8.9 µM BA 

SENGUPTA and SEN (1988) Corm 4.5 µM 2,4-D, and 

15% coconut milk 

89


Table 2.8: Corm induction treatments for various genera in Iridaceae. Details on the media 

modifications, temperature and the hours of light (Photoperiod) during corm induction is 

included. The period it took for corms to form is also given in months. The genera are grouped 

phylogenetically, with vertical text on the right showing classification. A question mark 

indicates that the specific parameter is not included in the described publication. 

Subfamily 

Ixioideae 

Tribe 

Ixieae 

Subtribe 

Romuleinae 

Ixiinae 

Gladiolinae 

Genus 

Crocus 

Ixia 

Dierama 

Sparaxis 

Gladiolus 

PLESSNER, ZIV and 

NEGBI (1990) 

Author(s) 

HOMES, LEGROS and 

JAZIRI (1987) 

Media 

modifications 

1.4 µM GA3; 2.3 

M 2,4-D; 2.3 M 

kinetin 

0 or 0.4 M 

kinetin and 4.5 or 

9.1 M 2,4-D; 2% 

sucrose 

Temperature 

Light (h) 

Period 

(months) 

15-25°C ? 3.6 

30°C 0 2 

SUTTER (1986) 0.3 M NAA 25°C 24 >2 

MADUBANYA (2004) 17.0 to 34.0 M 

PP3 

25°C 16 3 

HUSSEY (1976) 2% sucrose 20°C 16 6 -10 

DANTU and BHOJWANI 

(1987) 

DANTU and BHOJWANI 

(1995) 

DE BRUYN and 

FERREIRA (1992) 

GINZBURG and ZIV 

(1973) 

10% sucrose 25°C 24

Subfamily 


Iridoideae 

Tribe 


Watsonieae 

Irideae 

Tigrideae 

Subtribe 

Gladiolinae (continued) 

Freesiinae 

Hesperanthinae 

Tritoniinae 

(Unknown) 

Iridinae 

Cipurinae 

Genus 

Gladiolus (continued) 

Freesia 

Shizostylis 

Tritonia 

Watsonia 

Iris 

Cipura 

Author(s) 

Media 

modifications 

SEN & SEN (1995) 4.4 µM BA or no 

PGR's 

STEINTZ and LILIEN- 

KIPNIS (1989) 

STEINITZ, COHEN, 

GOLDBERG and 

KOCHBA (1991) 

34.0 or 170.2 µM 

PP3; 3%, 6% 

sucrose 

Temperature 


Light (h) 

Period 

(months) 

23 to 25°C 16 3 

25/20°C 

(day/night) 

6% sucrose 25/20°C 

(day/night) 

16 1 

16 1 

SUTTER (1986) 4.4 µM BA 25°C 24 2 

TAN NHUT, TEIXEIRA 

DA SILVA, HUYEN and 

PAEK (2004) 

1.0 M BA 15 or 20°C 24 2 

ZIV (1989) 1/2 strength MS; 

3.4 µM PP3 

25°C 24 4 to 5 

ZIV, RONEN and RAVIV 

(1998) 

8.7 M PP3 24/22°C 

(day/night) 

16 2.5 - 

3.5 

HUSSEY (1976) 2% sucrose 20°C 16 12 -14 

HUSSEY (1976) 2% sucrose 20°C 16 6-10 

ASCOUGH, SWART, 

FINNIE and VAN 

STADEN (2011) 

ASCOUGH, ERWIN, 

VAN STADEN (2008) 

none 10 or 15°C 16 3 

6% sucrose; 1 mg 

l-1 GA 3 

20°C 24 3 

HUSSEY (1976) 2% Sucrose 20°C 16 6 -10 

SENGUPTA and SEN 

(1988) 

1/2 strength MS; 

1% sucrose 

22-25°C 16 1 

91

3 Investigation into the habitat of Romulea sabulosa and 

Romulea monadelpha: Soil sampling and analysis 

3.1 INTRODUCTION 

R. sabulosa, one of the most attractive species of this genus, occurs west of 

Nieuwoudtville on sandy soil and in renosterveld on clay (MANNING & GOLDBLATT, 

1997; 2001). In these areas of renosterveld populations of R. monadelpha, another 

attractive species, can also be found. Other species of Romulea used in this study 

also occur in this area (Figure 2.1). 

In a study by KÖK et al. (2007) soil samples were taken during vegetative and 

flowering stages in the habitat of R. columnae in Turkey. This species is widely used 

as an ornamental plant (KÖK et al., 2007). The climate of this study area is 

somewhat similar to Namaqualand, with it being described as having a humid 

Mediterranean climate and a xeric period of 4 months. Analyses done on soil 

samples by KÖK et al. (2007) determined soil texture, pH, salinity and percentage 

nitrogen, phosphorous, potassium, organic matter and CaCO3. 

During a visit to the Nieuwoudtville Wildflower Reserve, Mr. Eugene Marinus showed 

me plants of R. monadelpha and R. sabulosa which he had grown from seed. 

Although not open, flower buds were visible on these plants and the plants appeared 

healthy and mature. He also mentioned that some of these plants had flowered in 

previous seasons. I enquired more about his methods and he subsequently showed 

me the soil he had used and allowed me to take samples. He had used a 1:1 mixture 

of soil from two locations about 20 m from each other. 

Observations of plants in natural conditions will be useful to compare with the 

morphology and size of the plants propagated by in vitro techniques. Soil texture, pH, 

salinity and percentage nitrogen, phosphorous and potassium was expected to be 

similar to that obtained by KÖK et al. (2007) in Turkey. 

92

Soil sampling and analysis 

The aims of this Chapter were to determine and measure the ecological variables in 

the soil of these plants so that these variables can be compared with observations 

made in vitro and ex-vitro for the same species in subsequent Chapters. 

3.2 MATERIALS AND METHODS 

Soil samples were taken from the Nieuwoudtville Wildflower Reserve (19° 8’ E, 31° 

24’ S). Soil samples were divided into two groups based on locality (the two soils 

used by Mr. Marinus). Sample 1 was described by Mr. Marinus as dolerite soil and 

sample 2 as tillite soil. 

Soils were analyzed by KwaZulu-Natal Department of Agricultural and Environmental 

Affairs Soil Fertility and Analytical Services, Pietermaritzburg. A texture test with 3 

fractions produced the amount of sand (0.02 - 2 mm), fine silt (0.02 - 0.002 mm) and 

clay (


The first number of the assessment has a value from 1 to 3; 1 being non-saline, 2 

potentially saline and 3 saline. The second number has a value from 4 to 6; 4 being 

non-sodic, 5 potentially sodic and 6 sodic. A sodic soil is defined as a soil with a low 

soluble salt content and a high exchangeable sodium percentage (ESP), with a usual 

ESP > 15 (SOIL CLASSIFICATION WORKING GROUP, 1991). A code 1,4 soil is 

described as suitable for irrigation, a code 2,5 soil as poorly drained and not suitable 

for irrigation and a code 3,6 soil as not suitable for irrigation. 

A pH test and a water content test were done on both samples. The pH of three 

different replicate samples was measured. Soil was mixed 1:1 (v/v) with water and 

homogenized with a stirrer apparatus. After about an hour the soil samples were 

vacuum filtered through Whatman No.1 filter papers and their pH was measured. The 

water content was calculated by weighing the soil, placing the soil in a drying oven 

set at 110°C and then subtracting the mass after no weight loss could be observed 

from the initial weight. 

3.3 RESULTS 

The colour of the two samples suggests that the drainage conditions of sample 1 is 

superior to that of sample 2 (DONAHUE et al., 1983) (Figure 3.1). Sample 1 also had 

more leaf and root material than sample 2, suggesting that it has more organic 

matter, and therefore a higher nutrient content, than sample 2 (DONAHUE et al., 

1983). It is notable that sample 2 appears more aggregated and dense. 

94

Figure 3.1: The colour and structure of samples 1 and 2. Horizontal bar = 20 mm. 


The texture test revealed that sample 1 is a clay soil due to its high content of 

particles smaller than


Table 3.1: Analysis results for two soil samples from the Nieuwoudtville Wildflower Reserve 

(19° 8’ E, 31° 24’ S). 

Water 

Total N 

Test Test parameters Sample 1 Sample 2 

Texture 

pH 

Fertility density 

Inorg. N 

Salinity 

content 

& C 

Sand % 31 63 

Fine Silt % 16 18 

Clay % 54 19 

Wet weight (g) 439.3 392.0 

Dry weight (g) 393.5 384.7 

Water content (%) 10.4% 1.9% 

Measured with fertility density 6.66 4.94 

Measured with soil salinity 7.30 4.99 

Personal test 7.81±0.05 6.72±0.5 

Sample density (g/ml) 1.13 1.23 

Phosphorous (%) 0.0003 0.0022 

Potassium (%) 0.0337 0.0246 

Calcium (%) 0.8796 0.0681 

Magnesium (%) 0.1587 0.0264 

Exchangeable acidy (cmol/l) 0.06 0.05 

Total cations 57.88 6.25 

Zinc (%) 0.0004 0.0006 

Manganese (%) 0.0002 0.0003 

Copper (%) 0.0003 0.0002 

MIR clay (%) 19 12 

MIR organic Carbon (%)

3.4 DISCUSSION 


R. culumnae grows in soils that are richer in nutrients and less saline compared to 

the soils in which R. monadelpha and R. sabulosa grows. The concentrations of 

nitrogen, phosphorous and potassium was higher in soil sampled from the natural 

habitat of R. culumnae in Turkey (KÖK et al., 2007). The soil of the habitat of R. 

culumnae had a smaller N:P:K ratio (1.000:6.897:1.069) and a lower salinity. The pH 

of the soils from the Namaqualand and Turkey is however similar. R. culumnae also 

occurs in a loamy clay soil, which is essentially a combination between the clay soil 

(sample 1) and a sandy loam soil (sample 2) in which R. monadelpha and R. 

sabulosa occurs, except for the slightly higher sand content that such a combination 

would have. 

Elements that are in abundance or that are lacking or low in the natural environment 

may have an effect on germination and growth in some species (BASKIN & BASKIN, 

1998). In soil sample 1 and 2 calcium is abundant. Although the percentage Fe 

content was not tested, the red colour of the clay indicates that the soils are rich in 

Fe. 

3.5 SUMMARY 

• The pH and soil texture of soils in which R. sabulosa and R. monadelpha, and 

R. culumnae occur is similar. 

• Soils in which R. sabulosa and R. monadelpha are found is more saline and 

has a much lower percentage nitrogen, phosphorous and potassium than soil 

in which R. culumnae is found. 

97

4 Germination physiology 


Species in the Iridaceae may exhibit morphological and/or morphophysiological 

dormancy. These mechanisms of delaying germination is due to the underdeveloped 

embryo of species in this family (TILLICH, 2003). The only literature published so far 

in relation to the germination of Romulea species is that of DENO (1993) and EDDY 

& SMITH (1975). These studies indicates that species of this genus requires a low 

temperature of 10°C for germination and temperatures between 16.5°C and 20°C 

may have an inhibitory effect on germination. EDDY & SMITH (1975) showed that 

application of KNO3 and pre-chilling for 5 days at 2°C did not increase germination of 

R. rosea. Seeds of this species are effectively dispersed by sheep and showed only 

38% germination in its faecal pad compared to the normal germination of 96% 

(EDDY & SMITH, (1975). This finding suggests that scarification is not an ideal 

treatment to break the dormancy of these seeds. R. rosea is an invasive species on 

other continents and some islands. This is the only Romulea species which shows 

high germination under natural environmental conditions. On the other hand, there 

are several other important potential horticultural species that are not easily 

germinated and their seed biology remains unknown. 

The aim of this Chapter was to understand the physiological mechanism behind 

dormancy and germination and to improve percentage germination of some 

important Romulea species with economic potential. Additionally, to test the 

germination mechanism of R. rosea in more detail, which can help in 

eradicating/controlling this species in countries where it is invasive. 


In South Africa, collection of many Romulea species is legally restricted due to the 

limited existence of natural populations. For this study, seeds were obtained from 

Silverhills Nursery, Kenilworth and African Bulbs, Napier. Both are South African 

98

Germination physiology 

companies. The availability of seeds of Romulea species at these nurseries was very 

limited and therefore only a small quantity could be purchased. Seeds of all Romulea 

species examined in this study were about one-year-old. To understand the basic 

physiological functions of these seeds, water content, imbibition rate and the viability 

of seeds both in vitro and ex vitro were determined. Seed surface and micropylar 

regions of these species were observed using a scanning electron microscope. 

4.2.1 Viability tests 

2,3,5-Triphenyl tetrazolium chloride (TTC) and embryo excision tests were conducted 

to examine seed viability. For the TTC test, the seeds were soaked in 1% TTC 

solution for one week in a glass vial and kept under constant dark conditions at 25 ± 

0.5°C. Subsequently, these seeds were cut into two and the percentage and degree 

of staining of the embryo and endosperm portions were recorded (INTERNATIONAL 

SEED TESTING ASSOCIATION, 1999). This test had four replicates of 10 seeds 

each. 

To excise the embryos, the seeds were first surface-decontaminated with 1.75% 

sodium hypochlorite solution with a few drops of Tween 20 for 15 min. Thereafter 

they were rinsed three times with sterile distilled water (ASCOUGH et al., 2007). 

Decontaminated seeds were placed in an autoclaved Petri dish with sterile distilled 

water. The seeds were left to imbibe in the laminar flow cabinet for 4 to 5 days. For 

embryo excision, slices of testa and endosperm were separated from the seed using 

a scalpel with a sterile blade, until the embryo was visible. Small pieces of 

endosperm around the embryo were carefully removed with a dissection needle. The 

embryo was then lifted from the endosperm with the help of a sterile dissecting 

needle and placed in an autoclaved Petri dish, filled with sterile distilled water to rinse 

off residual pieces of endosperm. Embryos were then placed into 33 ml culture tubes 

using an autoclaved Pasteur pipette filled with sterile distilled water. These tubes 

contained 10 ml Murashige and Skoog (MS) medium (MURASHIGE & SKOOG, 

1962) supplemented with 100 mg.l -1 myo-inositol and 3% sucrose, with pH adjusted 

to 5.7 and solidified with 0.8% agar. The experiment had four replicates of 5 seeds 

for each species. The culture tubes were sealed with Parafilm and placed under cool 

white fluorescent tubes (Osram L75 W/20X) with a 16-h photoperiod of 30.1 ± 3.4 

µmol m -2 s -1 irradiance at 25 ± 0.5°C. 

99

4.2.2 Water content and imbibition rate 


Water content was determined by weighing seeds before and after placing them in a 

drying oven set at 110°C. The weighing continued for six weeks until there was no 

further loss in seed weight. Per species, four replicates of 5 seeds each were used. 

The imbibition rate was determined by weighing the seeds after imbibing them for 3, 

6, 24, 48, 150, 168, 216, 264 and 336 h. Seeds were placed in 6.5 cm disposable 

plastic Petri dishes with filter paper (Whatman No.1) moistened with 3.5 ml of distilled 

water and subsequently kept moist by adding 1 to 3 ml of distilled water when 

needed. At each time interval, the seeds were removed from the Petri dishes, blotted 

dry, weighed and replaced in the respective Petri dishes. This experiment was 

conducted at room temperature (25 ± 0.5°C) using four replicates of 5 seeds for each 

species. 

4.2.3 Scanning electron microscopy 

Seeds were rinsed in 70% ethanol for 30 seconds and then allowed to dry on a paper 

towel. Seeds were mounted on 12 mm stubs and the micropylar regions and 

surfaces viewed using a scanning electron microscope (Zeiss EVO LS15 variable 

pressure). 

4.2.4 Ex vitro germination experiments 

Experiments were conducted to test the effect of temperature and light, cold and 

warm stratification, acid scarification, mechanical scarification, plant growth 

promoting substances and deficiency of nitrogen, phosphorous and potassium on 

germination of different Romulea species. 

If not stated otherwise, a growth chamber set at 20°C with a 16 h light: 8 h dark 

photoperiod and a photosynthetic photon flux density (PPFD) of 30.1 ± 4.0 µmol m -2 

s -1 provided by cool-white fluorescent lamps was used. Seeds were decontaminated 

by placing them in 0.2% mercuric chloride for 5 min and rinsing them once with tap 

water and twice with distilled water. Unless otherwise stated, seeds were placed in 

6.5 cm disposable plastic Petri dishes with two circles of filter paper (Whatman No.1) 

moistened with 3.5 ml of distilled water and test solutions. The filter paper was kept 

moist with the respective solutions when needed. In all germination experiments 10 

seeds were placed in each Petri dish with 4 replicates per treatment. All germination 

experiments were terminated after 90 days. Germination was recorded every second 

day and was considered complete when the radicle had emerged up to 2 mm. Seeds 

100


subjected to dark treatments were inspected every second day under a green “safe 

light” with a PPFD of 0.3 mol m −2 s −1 . Mean germination time (MGT) was calculated 

using the equation: MGT = (n×d)/N, where ‘n’ is the number of seeds germinated 

on each day, ‘d’ is the number of days the experiment has been active and ‘N’ is the 

total number of seeds germinated after 90 days (ELLIS & ROBERTS, 1981). 

To test the effect of temperature and light on germination of four Romulea species, 

decontaminated seeds were placed in growth chambers set at 10, 15, 20, 25, 30 and 

30/15 C (day/night) with 16 h light: 8 h dark photoperiod. Petri dishes were covered 

with aluminium foil for the dark treatment and were only opened under a ‘safe light’ 

as indicated earlier. 

For the stratification experiment, surface decontaminated seeds were placed 

between two layers of paper towelling moistened with distilled water. The towels 

were placed inside a plastic bag and were covered with aluminium foil to eliminate 

light. Seeds of all species were placed in a refrigerator at 5 C, except those of R. 

rosea which were placed at 30 C, as EDDY & SMITH (1975) reported pre-chilling did 

not improve the germination of this species. Seeds were kept under these conditions 

for 7, 14 and 21 days before placing them for germination in a growth chamber set at 

20 C. Control seeds did not receive stratification treatment and were placed in the 

same growth chamber with the seeds of the 7 day stratified set. 

For acid scarification, seeds were placed in 50% sulphuric acid for 5 min. Seeds 

were also mechanically scarified by rubbing them between two pieces of sand paper 

(P120 grade) for 1 min. Control seeds were not subjected to scarification treatment. 

To test the effect of various plant growth promoting substances, seeds of R. rosea 

were placed in Petri dishes with filter papers moistened separately with 10 -5 M of GA3, 

Kinetin, KNO3, IBA, NAA and IAA. Smoke water concentration of 1:500 (v/v) and 

butenolide (3-methyl-2H-furo[2,3-c]pyran-2-one) solution of 10 -8 M were also used in 

this study. Butenolide is a compound isolated from plant-derived smoke water that 

enhances germination of many seeds, including Eucomis (KULKARNI et al., 2006). 

Plant-derived smoke water was prepared according to BAXTER et al. (1994) and 

butenolide was isolated by the method of VAN STADEN et al. (2004). The effect of 

macro-nutrient deficiency was investigated by placing seeds on filter paper 

moistened with 3.5 ml of 50% Hoagland’s nutrient medium deficient of either nitrogen 

101


(-N), phosphorous (-P) or potassium (-K) respectively. The seeds that were 

germinated with 50% Hoagland’s medium consisting of all nutrients and with only 

distilled water served as controls. The filter paper was kept moist with these solutions 

throughout the duration of the experiment to maintain the levels of nutrients. 

4.2.5 In vitro germination experiments 

Seeds were decontaminated in accordance with the methods of ASCOUGH et al. 

(2007). Seeds were placed in 1.75% sodium hypochlorite solution with a few drops of 

Tween 20 for 15 min, after which they were rinsed three times with sterile distilled 

water. All seeds were placed in 33 ml culture tubes with 10 ml 1/10 strength MS 

media supplemented with 100 mg.l -1 myo-inositol and no sucrose (ASCOUGH et al., 

2007). Tubes were placed in a growth chamber set at 15°C and a 16 h light: 8 h dark 

photoperiod with PPFD of 30.1 ± 4.0 µmol m -2 s -1 . Four replicates of 10 seeds (10 

tubes) each of R. autumnalis, R. citrina, R. cruciata, R. diversiformis, R. flava, R. 

leipoldtii, R. minutiflora, R. pearsonii, R. sabulosa and R. tabularis were used. 

Considering R. sabulosa’s commercial potential, additional experiments were 

conducted by placing four replicates of 50 seeds each at 15 and 20°C. 

4.2.6 Statistical analysis 

Percentage germination data were arcsine transformed and analysis of variance 

(ANOVA) was calculated. Differences between means of percentage germination 

were tested with Least Significant Difference (LSD) at the 5% level. GENSTAT ® 

Release 4.21 statistical package was used for analysis. 

102

4.3 RESULTS 

4.3.1 Viability tests 


The results of the TTC and embryo excision tests are shown in Table 4.1. According 

to the TTC test, seed embryo and endosperm of R. camerooniana and R. flava 

showed < 61% staining indicating low to moderate viability respectively. All other 

tested species of Romulea exhibited over 90% staining of embryo and endosperm 

indicating high seed viability. The embryo excision test of R. camerooniana and R. 

rosea did not show any in vitro response. An in vitro response was however 

observed for all other species. Excised embryos of R. diversiformis and R. leipoldtii 

showed the highest percentage in vitro response. TTC and embryo excision tests 

were comparable for species R. diversiformis, R. leipoldtii and R. monadelpha (Table 

4.1). The results of both tests differed for R. flava, R. minutiflora and R. sabulosa. 

Table 4.1: Seed viability tests of different Romulea species. 

Species 

TTC (embryo + endosperm) 

(% staining ) 

Embryo excision 

(% response) 

R. camerooniana 45 ± 19 0 ± 0 

R. diversiformis 100 ± 0 100 ± 0 

R. flava 60 ± 13 95 ± 5 

R. leipoldtii 95 ± 6 100 ± 0 

R. minutiflora 92 ± 12 70 ± 30 

R. monadelpha 100 ± 0 90 ± 10 

R. rosea 90 ± 6 0 ± 0 

R. sabulosa 95 ± 5 54 ± 11 

4.3.2. Water content and imbibition rate 

R. leipoldtii seed had the highest initial water content (37 ± 1.4%) followed by R. flava 

(30 ± 2.1%). The lowest seed water content was determined for R. camerooniana 

and R. diversiformis (23%). Overall the range of seed water content was between 23 

and 37% in the studied Romulea species (Figure 4.1). In most of the species the rate 

of imbibition sharply increased in the first three hours and was gradual up to 6 h, 

followed by a sharp increase in water uptake up to 48 h, after which it was slow. R. 

103


rosea seeds however imbibed much more water between 3 h and 6 h and had a 

higher percentage water content at 6 h (89 ± 7.3 %) than any other of the species 

tested (Figure 4.1). 

% water 

% water 

% water 

% water 

200 

180 

160 

140 

120 

100 

80 

60 

40 

20 

0 

200 

180 

160 

140 

120 

100 

80 

60 

40 

20 

0 

200 

180 

160 

140 

120 

100 

80 

60 

40 

20 

0 

200 

180 

160 

140 

120 

100 

80 

60 

40 

20 

Romulea camerooniana Romulea diversiformis 

Hours of imbibition 

Romulea flava 

0 

0 3 6 24 48 120 168 216 264 336 

Romulea leipoldtii 

Romulea minutiflora Romulea monadelpha 

Romulea rosea R. sabulosa 

0 3 6 24 48 120 168 216 264 336 

Hours of imbibition 

Figure 4.1: Water content (value at day zero) and imbibition rates of seeds of eight 

species of Romulea. Error bars indicate standard error of the mean. 

104

4.3.3. Scanning electron microscopy 


Figure 4.2: Scanning electron microscopic images of seeds arranged from the smallest to 

the largest for size comparison. Romulea leipoldtii (A); R. flava (B); R. minutiflora (C); R. 

sabulosa (D); R. camerooniana (E); R. rosea (F); R. diversiformis (G); R. monadelpha (H). 

Horizontal bar = 1 mm. 

Figure 4.3: Scanning electron micrographs of the seed surfaces of Romulea camerooniana (A); 

R. diversiformis (B); R. flava (C); R. leipoldtii (D); R. minutiflora (E); R. monadelpha (F); R. rosea 

(G) and R. sabulosa (H). Horizontal bar = 10 µm (the same magnification was used for all 

species). 

105


Figure 4.4: Scanning electron micrographs of the micropylar regions of seeds of Romulea 

camerooniana (A); R. diversiformis (B); R. flava (C); R. leipoldtii (D); R. minutiflora (E); R. 

monadelpha (F); R. rosea (G) and R. sabulosa (H). Horizontal bar = 20 µm. 

The size and shape of seeds of different Romulea species showed large variations 

as seen in Figure 4.2. Both these parameters have a significant influence on water 

imbibition. R. rosea had the roughest seed surface, followed by R. leipoldtii and R. 

diversiformis (Figure 4.3). The seed surface of R. flava appears to be the smoothest, 

followed by R. sabulosa and R. minutiflora. In Romulea species, the sizes of the 

micropylar region are correlated to seed size with R. monadelpha and R. 

diversiformis having the largest micropylar regions (Figure 4.4). The micropylar 

regions of R. minutiflora and R. rosea appear more membranous than those of other 

species. The micropylar region of R. sabulosa seeds appears to be the densest, 

followed by R. camerooniana. 

106

Table 4.2: Effect of different treatments on seed germination of four Romulea species. Asterisk (*) indicates seed germination under 16 h photoperiod at 

20 ± 0.5°C. The number sign (#) indicates that the seeds initiated germination during stratification. 

R. diversiformis R. flava 

Species 

R. monadelpha R. rosea 

Treatment 

Temperature (°C) 

(16 h photoperiod) 

Germination (%) MGT Germination (%) MGT (days) Germination (%) MGT (days) Germination (%) MGT (days) 

10 65 ± 0.9 a 41 ± 3 b 0 ± 0 d 0 ± 0 d 0 ± 0 b 0 ± 0 b 90 ± 1 a 32 ± 1 e 

15 0.5 ± 0.3 d 30 ± 6 b 5 ± 1d 57 ± 0 a 0 ± 0 b 0 ± 0 b 68 ± 1 b 72 ± 2 b 

20 0 ± 0 d 0 ± 0 c 0 ± 0 d 0 ± 0 d 3 ± 1 b 42 ± 0 a 15 ± 1 c 57 ± 1 c 

25 0 ± 0 d 0 ± 0 c 0 ± 0 d 0 ± 0 d 0 ± 0 b 0 ± 0 b 0 ± 0 d 0 ± 0 f 

30 0 ± 0 d 0 ± 0 c 0 ± 0 d 0 ± 0 d 0 ± 0 b 0 ± 0 b 0 ± 0 d 0 ± 0 f 

30/15 

(Constant dark) 

0 ± 0 d 0 ± 0 c 0 ± 0 d 0 ± 0 d 0 ± 0 b 0 ± 0 b 0 ± 0 d 0 ± 0 f 

10 60 ± 1.5 ab 39 ± 1 b 8 ± 5 d 32 ± 1 b 3 ± 1 b 64 ± 0 a 95 ± 1 a 36 ± 1 e 

15 48 ± 0.6 bc 41 ± 4 b 18 ± 5 c 53 ± 3 a 23 ± 1 a 43 ± 4 a 5 ± 1 d 57 ± 2 c 

20 0 ± 0 d 0 ± 0 c 0 ± 0 d 0 ± 0 d 8 ± 1 b 56 ± 0 a 5 ± 1 d 76 ± 0 a 

25 0 ± 0 d 0 ± 0 c 0 ± 0 d 0 ± 0 d 0 ± 0 b 0 ± 0 b 0 ± 0 d 0 ± 0 f 

30 0 ± 0 d 0 ± 0 c 0 ± 0 d 0 ± 0 d 0 ± 0 b 0 ± 0 b 0 ± 0 d 0 ± 0 f 

30/15 

Cold stratification (5°C)* 

0 ± 0 d 0 ± 0 c 0 ± 0 d 0 ± 0 d 0 ± 0 b 0 ± 0 b 0 ± 0 d 0 ± 0 f 

(Days) Warm stratification (30°C) 

0 0 ± 0 d 0 ± 0 c 0 ± 0 d 0 ± 0 d 0 ± 0 b 0 ± 0 b 5 ± 1 d 77 ± 7 a 

7 0 ± 0 d 0 ± 0 c 0 ± 0 d 0 ± 0 d 3 ± 1 b 52 ± 0 a 5 ± 1 d 62 ± 2 cd 

14 18 ± 0.5 d 10 ± 5 c 43 ± 5 b 10 ± 6 c 0 ± 0 b 0 ± 0 b 70 ± 2 b 64 ± 2 c 

21 38 ± 0.9 c # 1 ± 0 c 53 ± 9 a # Scarification* 

1 ± 0 d 0 ± 0 b 0 ± 0 b 73 ± 2 b 65 ± 1 c 

Control 0 ± 0 d 0 ± 0 c 0 ± 0 d 0 ± 0 d 0 ± 0 b 0 ± 0 b 0 ± 0 d 0 ± 0 f 

Acid (50% H2SO4) 63 ± 0.8 ab 71 ± 3 a 0 ± 0 d 0 ± 0 d 0 ± 0 b 0 ± 0 b 0 ± 0 d 0 ± 0 f 

Mechanical (sand paper) 0 ± 0 d 0 ± 0 c 0 ± 0 d 0 ± 0 d 0 ± 0 b 0 ± 0 b 0 ± 0 d 0 ± 0 f 

Means (± SE) in the column with different letters are significantly different according to LSD at the 5% level (P < 0.05) 

107

Germination (%) 

50 

40 

30 

20 

10 

0 

77 

Control (water) 

68 

Control (HS 50%) 

Treatment 

4.3.4. Ex vitro germination experiments 

62 

-N 

64 

-P 

87 

-K 

GA3 (10 -5 M) 

65 

Kinetin (10 -5 M) 

KNO 3 (10 -5 M) 

* * 

IBA (10 -5 M) 

NAA (10 -5 M) 

IAA (10 -5 M) 

Smoke water (1:500) 


Butenolide (10 -8 M) 

Figure 4.5: Effect of nutrients without N, P or K, plant growth promoting substances and 

smoke constituents on seed germination of Romulea rosea under 16 h photoperiod at 20 

± 0.5°C. A number above the standard error bar represents mean germination time and an 

asterisk denotes that the treatment was significantly different from the control (water) 

according to LSD test at the 5% level. 

Germination was observed for the four tested Romulea species. R. diversiformis 

seeds showed the highest percentage germination when placed at 10°C under 

alternating light (16 h photoperiod), in the constant dark conditions and with seed 

scarification treatment (Table 4.2). These results were significantly different from 

other treatments for R. diversiformis, with the exception of constant dark at 15°C, 

where it did not differ significantly in some cases. Although acid scarification 

66 

67 

68 

68 

68 

108


exhibited higher percentage germination, the MGT was significantly greater than for 

the other treatments. Cold stratification of seeds for 14 and 21 days significantly 

increased percentage germination in comparison to the non-stratified seeds with 

significantly shorter MGT (some seeds germinated during stratification period) (Table 

4.2). 

R. flava seeds showed some germination both under 16 h photoperiod and constant 

dark conditions at 15°C. However, 14 and 21 days of stratification at 5°C of seeds 

achieved significantly greater percentage germination compared to all other 

treatments examined. Twenty-one day stratification had the shortest MGT of only one 

day due to the incidence of germination during the treatment period (Table 4.2). 

In the case of R. monadelpha, placing seeds at 15°C in the dark was the only 

treatment that significantly increased germination compared to the other treatments 

(Table 4.2). 

R. rosea seeds had significantly higher percentage germination when placed at 10°C 

both under 16 h photoperiod and constant dark in comparison to other treatments. 

Mean germination time was also significantly shorter for these treatments (Table 4.2). 

Warm stratification did not significantly increase percentage germination in 

comparison to the seeds incubated at 10°C. Seed germination of this species at 20- 

35°C was very low. These temperatures are generally favourable for many weed 

species, seeds of R. rosea were therefore tested for nutrients, plant growth 

promoting substances and smoke components at only 20 ± 0.5°C. In this experiment, 

seeds that were treated with KNO3 and IBA solutions of 10 -5 M yielded significantly 

higher percentage germination over the control (water) (Figure 4.5). Percentage 

germination at HS 50% (all nutrients), -K, kinetin, NAA, IAA and butenolide 

treatments showed an increase compared to control (water). However, these results 

were not significantly different to the control (water). No germination was recorded for 

GA3 and smoke-water-treated seeds. 

109

Germination (%) 

100 

80 

60 

40 

20 

0 

cd 

Romulea autumnalis 

d 

Romulea camerooniana 

bcd 

Romulea citrina 

Romulea cruciata 

a a 

Romulea diversiformis 

abc 

Romulea flava 

Species 

Romulea leipoldtii 

ab 

Romulea multiflora 

abcd 

d d cd 

Figure 4.6: In vitro seed germination of different Romulea species at 15°C after 2 

Romulea pearsonii 


months. Standard error bars with different letters are significantly different according to 

LSD at the 5% level. 

4.3.5. In vitro germination experiments 

R. diversiformis, R. flava, R. leipoldtii and R. minutiflora showed the highest 

germination (> 55%) in comparison to other species, although in some cases these 

results were not significant (Figure 4.6). The lowest germination was recorded for R. 

camerooniana, R. cruciata, R. sabulosa and R. tabularis. Only few seeds of R. 

sabulosa germinated (9 out of 200 seeds) when incubated at 15°C, with no 

germination at 20°C. Seeds did not geminate at 15°C and 20°C in the case of R. 

camerooniana. 

Romulea sabulosa 

Romulea tabularis 

110



A short rainy season plays an important role for the wild flowers of Namaqualand. 

Plants must complete their life-cycle during this period, ensuring successful 

germination and subsequent establishment due to sufficient moisture. Even under 

moist environmental conditions there may be constrains on germination to the seeds 

that are on the soil surface compared to those seeds that are buried and which form 

part of the soil seed bank (MEEKLAH, 1958; EVANS et al., 1967). This can be 

attributed to high fluctuations in moisture and humidity which results in unfavourable 

conditions for germination (MILLER & PERRY, 1968; DOWLING et al., 1971). Seed 

surface characteristics in modifying the seed/soil interface has been emphasized by 

SEDGLEY (1963) and HARPER AND BENTON (1966). These workers showed that 

germination was promoted when a greater area was in contact with the substrate. 

Hence size, surface and micropylar region are important factors of the seed to be 

considered for the process of water-uptake. 

Many Romulea species have a narrow and limited distribution and germination plays 

a significant role in their survival. It is therefore necessary to understand the seed 

structure of different Romulea species and their ecological relevance. In comparison 

to other species, high initial water content and fast imbibition of R. leipoldtii seeds 

can be attributed to its small seed size. Large surface area to volume ratio requires 

the seed to have more water reserves (as it is more prone to desiccation) and would 

allow the seed to absorb more water in a shorter time (when available). The small 

seed size of R. flava also requires it to have more water reserves to avoid 

dehydration. These species showed the roughest seed surface after R. rosea, R. 

leipoldtii and R. diversiformis, which further increase surface area and the risk of 

dehydration (but also allows it to imbibe more water in a shorter time). R. leipoldtii 

and R. flava are all found in areas with highly variable rainfall (Figure 2.21). 

The seeds of R. camerooniana are smooth and their seeds are larger than those of 

R. leipoldtii and R. flava, resulting in a small surface area to volume ratio. These 

seeds are however, small enough to have a relatively high imbibition rate because of 

the resulting larger surface area to volume ratio. This species also occur in areas 

with higher and more consistent rainfall than other species of this genus (Figure 2.7). 

The seeds of R. camerooniana therefore do not require a high initial water content. 

111


The relatively low initial water content and slow water imbibition of R. monadelpha 

seeds can be attributed to their large size and smooth surface compared to the 

seeds of other Romulea species, while the relatively low water content of R. sabulosa 

seeds can be attributed to their relatively smooth seed surface and compact 

micropylar region. 

It is interesting to note that the eight species used in water content and imbibition 

experiments can be divided perfectly into their subgenera by looking at variability of 

their initial water content, as species in the subgenus Romulea (R. camerooniana, R. 

flava, R. leipoldtii and R. minutiflora) all had variable initial water content when 

compared to all studied species in the subgenus Spatalanthus (R. diversiformis, R. 

monadelpha, R. rosea and R. sabulosa). It was also noted that seed sizes of species 

in the subgenus Romulea had more variability compared to species in the subgenus 

Spatalanthus, which is a possible explanation for the variability in water content. 

The rough seed surface and membranous micropylar regions of R. rosea, enables 

them to absorb a relatively large amount of water; they also initiated an increase in 

imbibition rate 3 h before seeds of other species. R. minutiflora has a smooth seed 

surface, compact micropylar region and small seed size to facilitate its low water 

absorption capacity. The larger the seed, the higher the relative water capacity and 

imbibition, these are possible explanations why R. rosea is a more successful 

invasive plant than R. minutiflora and other species. 

The seeds of R. diversiformis and R. rosea showed high germination both under 

alternating and constant dark conditions at 10°C. These findings suggest that these 

seeds are not specifically light or dark requiring. Percentage germination of R. 

monadelpha seed was significantly higher under constant dark conditions at 15°C 

than any other treatment examined, which indicates a possible negatively 

photoblastic nature. However, this species had very low germination and therefore 

more investigation on seed physiology is required which may help improving 

germination of this species. On the other hand, R. flava seeds did not respond 

effectively when subjected to different temperatures under both light and dark 

conditions. Interestingly, this species exhibited increasing percentage germination as 

the cold stratification period was prolonged. It was observed that these seeds 

germinated during the stratification period. This suggests that R. flava seeds require 

a wet cold winter season for germination. R. sabulosa seeds did not respond to ex 

vitro germination but little germination was recorded in in vitro experiments. The 

112


intact seed of these species may need a longer period of cold stratification, or culture 

conditions at 5°C. There is a period of about two months with average daily minimum 

temperatures of 5°C in Nieuwoudtville where R. sabulosa is found (Figure 2.20). 

Although R. diversiformis, R. flava, R. leipoldtii and R. minutiflora all had in vitro 

germination percentages higher than 57% at 15°C, seeds of other species showed 

very low germination. R. camerooniana was the only species of which neither the 

seeds germinated nor the embryos responded. Low viability of R. camerooniana 

seeds was indicated by TTC. However, all species endemic to South Africa showed 

positive responses to either seed or embryo germination treatments. 

The results of the temperature experiment for R. rosea confirms that of EDDY & 

SMITH (1975) who found that R. rosea seeds has a clear germination optimum in the 

temperature range of 9.5 to 13°C. These results suggest that invasive individuals of 

R. rosea could be eradicated (by mechanical or chemical control) when temperatures 

are low. EDDY & SMITH (1975) reported that pre-chilling of seeds or KNO3 treatment 

does not increase germination of this species. However, in the present study KNO3 

significantly increased germination over the control. This result was confirmed where 

seed germination was low in Hoagland’s nutrient solution without K in comparison to 

the control (HS 50%). These findings suggest that the use of fertilizers containing 

high potassium should be avoided to reduce R. rosea invasion which will be higher in 

agricultural areas. 

It appears that seeds of R. diversiformis, R. flava, R. leipoldtii, R. minutiflora, R. 

monadelpha, and R. sabulosa all exhibit non-deep endogenous morphophysiological 

dormancy, as excised embryos showed a growth response and the causes could 

include a physiological germination inhibiting mechanism or an underdeveloped 

embryo (BASKIN & BASKIN, 1998). The percentage germination of R. rosea seeds 

on filter paper moistened with KNO3 was significantly higher than the control, this 

response to potassium nitrate is a characteristic of non-deep morphophysiological 

dormancy (COPELAND, 1976). The causes of non-deep physiological dormancy 

include the physical barrier created by covering structures, the resulting low oxygen 

supply to the embryo, inhibitors within the covering structures and/or 

physical/chemical changes in the covering structures (BASKIN & BASKIN, 1998). 

COPELAND (1976) states that the dormancy of such seeds can often be broken by a 

cold stratification treatment. This happened in the case of R. flava. However, this was 

not tested for R. leipoldtii and R. minutiflora due to limitations in seed availability. The 

113


seeds of R. camerooniana appears to have deep endogenous morphophysiological 

dormancy, as excised embryos showed no growth response and the apparent 

causes may include a physiological germination inhibiting mechanism or rudimentary 

embryos (BASKIN & BASKIN, 1998). 

4.5 SUMMARY 

• Germination experiments revealed that R. diversiformis and R. rosea seeds 

germinate best at 10°C. R. flava seeds germinate best when cold stratified for 

21 days and R. monadelpha seeds showed some response at 15°C in the 

dark indicating photoblastically negative behaviour. 

• An initial in vitro germination experiment showed germination above 57% for 

R. diversiformis, R. leipoldtii, R. minutiflora and R. flava seeds placed at 15°C; 

while seeds of other species placed at 15°C had less than 30% germination. 

• Seeds of R. diversiformis, R. flava, R. leipoldtii, R. minutiflora, R. monadelpha 

and R. sabulosa may have non-deep endogenous morphophysiological 

dormancy, whereas seeds of R. camerooniana appear to have deep 

endogenous morphophysiological dormancy. 

114

5 In vitro culture initiation and multiplication 


Due to the low germination observed for some species of horticultural importance 

such as R. monadelpha and R. sabulosa, an in vitro culture initiation protocol was 

developed for these two species and some other species of interest. Developing a 

micropropagation protocol will be essential in the commercialization of these species 

as it enables high propagation rates (LILIEN-KIPNIS & KOCHBA, 1987; PIERIK, 

1997). Developing a micropropagation protocol for R. sabulosa may also aid in the 

conservation of this rare and beautiful flower, as it has been shown that 

micropropagation can play a significant role in plant conservation (WOCHOCK, 1981; 

SARASAN et al., 2006; SHIBLI et al., 2006; WITHERS, 2008). 

Explant selection is an important step in the micropropagation process. Explant type 

can affect culture survival, vigour, regenerative capacity, bulking time, contamination 

rates and health. Explant size, for instance, influences survival and contamination 

rates (KYTE & KLEYN, 1996; SMITH, 2000). Explant maturity also affects survival 

and contamination rates as well as regenerative capacity and bulking time (KYTE & 

KLEYN, 1996; SMITH, 2000). Explant source affects survival, regenerative capacity, 

contamination rates and health (SMITH, 2000; AHLOOWALIA & PRAKASH, 2002). 

The preparation of the explant and the culture initiation medium is equally important 

(SMITH, 2000). See Table 2.6 for a review on explant types and media used for 

Iridaceae direct shoot culture initiation. 

The main aim of this Chapter was to determine the most appropriate explant type 

and medium for R. camerooniana, R. diversiformis, R. flava, R. leipoldtii, R. 

minutiflora, R. monadelpha, R. rosea and R. sabulosa shoot culture initiation. Some 

embryos of R. sabulosa were also used in an experiment aimed to produce 

embryogenic tissue. Shoots of R. sabulosa were used in experiments aimed to 

identify physical and chemical stimuli to significantly increase shoot multiplication. 

115


In vitro culture initiation and multiplication 

Seeds were obtained from Silverhill Nurseries, Kenilworth, South Africa and African 

Bulbs, Napier, South Africa. Although using seedling organ explants is common for 

culture initiation in the Iridaceae, seeds of the attractive and vulnerable R. sabulosa 

did not germinate and some other attractive species had low germination. Because 

these experiments were done before the cold stratification experiments described in 

Chapter 4, germination was slow. In an attempt to obtain cultures of all species in a 

shorter time embryo rescue techniques were employed for eight species. 

If not stated otherwise, a MS medium supplemented with 100 mg.l -1 myo-inositol and 

3% sucrose, with pH adjusted to 5.7 and solidified with 0.8% agar was used. All 

experiments were conducted in a laminar flow hood and cultures were placed in a 

growth room set at 25°C under 4.3 µmol m –2 s –1 light using Osram ® 75 W cool white 

fluorescent tubes with a 16/8 light/dark photoperiod. The duration of all experiments 

was two months. Shoots multiplied for further experiments were subcultured onto the 

medium that produced the best shooting response of the initial explant every two 

months. 

5.2.1 Explants from seedlings 

Seedlings from initial in vitro germination experiments described in Chapter 4 were 

used for a small experiment to assess the responsiveness of the species of which 

seeds germinated. Only 5 replicates were used per treatment, as the availability of 

seeds of Romulea species was very limited at the time and therefore only small 

quantities could be purchased. 

Seedlings with stems longer than 30 mm were removed from the culture tubes using 

autoclaved forceps. The seedlings were then placed on Petri dishes and cut into 

three sections using a scalpel and blade. They were divided into shoot (>10 mm), 

hypocotyl (10 mm) and root (>10 mm) sections. For hypocotyl explants the remainder 

of the seed was removed. All seedling organs were placed in 33 ml culture tubes with 

10 ml MS media supplemented with various plant growth regulators. 

116


Seedling organs of R. flava and R. leipoldtii were placed in culture tubes with MS 

media with nine different plant growth regulator treatments; a control with no plant 

growth regulators, 4.4 µM BA, 22.2 µM BA, 4.5 µM 2,4-D, 22.6 µM 2,4-D, 4.4 µM BA 

and 4.5 µM 2,4-D, 4.4 µM BA and 22.6 µM 2,4-D, 22.2 µM BA and 4.5 µM 2,4-D, and 

22.2 µM BA and 22.6 µM 2,4-D. Seedling organs of R. diversiformis and R. 

minutiflora were placed in culture tubes with MS media with 12 different plant growth 

regulator treatments; a control with no plant growth regulators, 2.3 µM kinetin, 23.2 

µM kinetin, 5.4 µM NAA, 26.9 µM NAA, 53.7 µM NAA, 2.3 µM kinetin and 5.4 µM 

NAA, 23.2 µM kinetin and 5.4 µM NAA, 2.3 µM kinetin and 26.9 µM NAA, 23.2 µM 

kinetin and 26.9 µM NAA, 2.3 µM kinetin and 53.7 µM NAA, and 23.2 µM kinetin and 

53.7 µM NAA. 

5.2.2 Explants from embryos 

The seeds were surface sterilised as with in vitro germination experiments described 

in Chapter 4 and imbibed for 5 days. Decontaminated seeds were placed in a Petri 

dish with sterile distilled water. The seeds were left to imbibe in the laminar flow 

cabinet and were dissected every 24 h to test for ease of dissection. Seeds could 

only be dissected after 4 days, as the blade could not penetrate the periderm (fruit 

coat) with a clean cut with less time for imbibition. The embryos that were dissected 

after 7 days of imbibition did not show any growth response when placed on media. 

To excise the embryo some of the surrounding endosperm had to be cut away (See 

Figure 5.1). 

117


Figure 5.1: General embryo excision procedure for Romulea seeds. An outer view, as one 

would view it through a stereo microscope, as well as a view relative to the embryo is provided 

so that the importance of the placing of the incisions can be seen. Step 1 is viewed from the 

top, Step 2 is a side view, Steps 3 and 5 are bottom views 90° to the incision made in Step 2. 

Step 4 is a side view. 

In Step 1 the seed was gripped with a pair of fine forceps and two incisions were 

made. The excess tissue was scrapped to the front of the Petri dish. In Step 2 the 

seed was turned onto one of the flat sides created by the incision done in Step 1. 

When turning on the bottom light of the dissection microscope the light radiated 

through the seed, making the embryo visible. Another incision was then made below 

the embryo. The seed was then turned onto the Petri dish as indicated in Step 3 to 

check for embryo visibility and to allow a better cutting angle. Small slices of tissue 

could then be removed as indicated in Step 4 until the embryo was clearly visible 

through the somewhat transparent endosperm tissue. An autoclaved dissection 

needle was then used to carefully flake away the sections indicated in Step 5 (Figure 

5.1). The embryo was then lifted out of the endosperm tissue and placed in a Petri 

dish with sterile distilled water to rinse off residual pieces of endosperm. 

Excised embryos were placed into 33 ml culture tubes with 10 ml nutrient media 

using an autoclaved Pasteur pipette with sterilised distilled water immediately after 

excision. Treatments with various concentrations of plant growth regulators in the 

growth media were prepared for each species. 

118


R. diversiformis, R. flava, R. minutiflora and R. monadelpha embryos were placed on 

media with no plant growth regulators and media supplemented with 2.3, 4.7 and 

23.2 µM kinetin. For R. diversiformis and R. minutiflora 10 embryos were used for 

each treatment, while 20 embryos were used per treatment for R. flava and R. 

monadelpha. 

Embryos of R. camerooniana and R. rosea were placed on 13 media treatments, a 

medium with no plant growth regulators, media with 2.3, 4.7 or 23.2 µM kinetin or 

mTR and media supplemented with 2.3, 4.7 or 23.2 µM kinetin or mTR in 

combination with 0.5 M NAA. 

For R. sabulosa an experiment was designed to investigate the suitability of a set of 

PGR treatments for shoot culture initiation and embryogenesis. The experiment had 

12 treatments and was repeated twice, first with 20 replicates and then with 10 

replicates. The 12 treatments consisted of media supplemented with combinations of 

0, 2.3, 4.7 and 23.2 M kinetin, and 0, 2.2 and 4.5 M 2,4-D. An experiment with 3 

treatments, media supplemented with 4.4 M BA, 22.2 M BA and 2.3 M kinetin 

with 5.4 M NAA was also initiated. This experiment was only repeated once with 20 

replicates. This is because of the vitrification and abnormal growth of shoots cultured 

on media supplemented with BA. 

After two months the number of shoots formed per explant, the presence or absence 

of roots and the morphology was recorded. The cultures that appeared embryogenic 

where placed in culture bottles with 30 ml of MS media supplemented with 100 mg.l -1 

myo-inositol, 30 g.l -1 sucrose and 8 g.l -1 activated charcoal after 2 months. 

Culture morphology was not only observed with the naked eye, but also under a light 

microscope. Samples of tissue suspected to have embryo initials were prepared by 

Epon resin embedding, sectioned using a LKB Ultrotome II microtome, stained with 

Ladd’s multiple stain and viewed with an Olympus AX 70 stereo microscope. 

119

5.2.3 Explant comparison 


R. leipoldtii seeds were used, not only because of their high germination and in vitro 

shooting in preliminary seedling organ experiments, but also because this species 

has the smallest seeds. If embryo excision is possible for a species with such a small 

seed it shows that seed size is not a limitation to using embryo excision in a culture 

initiation protocol for Romulea species. 

Seeds (200) were surface sterilised and germinated at 15°C as with in vitro 

germination experiments described in Chapter 4. After 2 months when sufficient 

seeds germinated, 130 seedling hypocotyls and 130 embryos were excised 

simultaneously. Seedlings with stems longer than 30 mm were then removed from 

tubes and placed on Petri dishes and cut into 3 sections using a sterilised scalpel 

and blade. For this experiment only the hypocotyl was used because of its higher 

percentage in vitro response. The seedling hypocotyls and embryos were then 

placed in 33 ml culture tubes with 10 ml nutrient media with 13 medium treatments, a 

medium with no plant growth regulators, media with 2.3, 4.7 or 23.2 µM kinetin or 

mTR and media supplemented with 2.3, 4.7 or 23.2 µM kinetin or mTR in 

combination with 0.5 M NAA. Ten replicate explants were used per treatment and 

the experiment was repeated twice. After two months the number of shoots formed 

per explant and the presence or absence of roots was recorded. 

5.2.4 Shoot multiplication 

R. sabulosa shoots initiated and multiplied with 23.2 M kinetin were placed in tubes 

containing MS media supplemented with three different kinetin concentrations (2.3, 

4.7 and 23.2 M) used for shoot initiation. These tubes were then placed at 25°C in a 

growth room with 16 hour light and a growth room with 24 hour light to test the effect 

of photoperiod on shoot multiplication rate. The effect of temperature was examined 

by placing tubes with shoots on MS media supplemented with 23.2 M kinetin in 

growth chambers with a 16/8 light/dark photoperiod set at a range of temperatures 

from 10°C to 30°C under 3.4 µmol m –2 s –1 light Osram ® 75 W cool white fluorescent 

tubes with a 16/8 light/dark photoperiod. These experiments were repeated twice and 

10 explants were used per treatment. After two months the number of shoots formed 

was recorded. 

120


Another experiment was conducted with R. sabulosa shoots to investigate the effect 

of other cytokinins and different concentrations of kinetin on shoot multiplication. R. 

sabulosa shoots initiated and multiplied with 23.2 M kinetin were placed in 250 ml 

jars containing 30 ml MS media supplemented with no plant growth regulators 

(control) or 2.5, 5.0, 7.5 M kinetin, BA, mT, mTR or MemTR. For each treatment, 5 

replicate bottles with 4 explants each were used. After 2 months the number of 

shoots formed was recorded. 

5.2.5 Statistical analysis 

Data were analyzed for significant differences by one-way analysis of variance 

(ANOVA), and means separated using either Tukey’s HSD test or Duncan’s multiple 

range test (DMRT) at 5% level of significance (P 0.05). Percentage data were 

converted to proportion, arcsine transformed and then analyzed. GenStat ® (VSN 

International, Hemel Hempstead, U.K.) version 11.1 statistical package was used to 

analyze the data. 

5.3 RESULTS 

5.3.1 Explants from seedlings 

In the preliminary experiment with R. diversiformis, R. flava, R. leipoldtii and R. 

minutiflora, no shoots developed from root and shoot explants and shoots only 

developed from seedling hypocotyls (Results not shown due to low replication). 

Hypocotyls of R. leipoldtii placed on a medium supplemented with 22.2 µM BA 

developed the largest number of shoots. R. flava seedling hypocotyls produced less 

shoots per explant than those of R. leipoldtii on the same medium. R. minutiflora 

hypocotyls only formed shoots when placed on a medium supplemented with 23.2 

µM kinetin and 5.4 µM NAA. R. diversiformis hypocotyls formed no shoots and only 

developed abnormal root-like structures. The shoots of R. leipoldtii appeared vitrified 

and when 50 shoots, multiplied on a medium supplemented with 23.2 µM kinetin, 

were placed in 33 ml culture tubes with 10 ml of MS media with no plant growth 

regulators, no rooting was observed after 2 months. 

121


R. flava and R. leipoldtii produced callus with shoot initials on the media treatments 

with the highest concentrations of both cytokinins and auxins. These shoot initials 

and the shoots formed after one month from them, when transferred MS media with 

no plant growth regulators, also appeared vitrified and no roots or root initials were 

observed after 2 months. 

5.3.2 Explants from embryos 

R. diversiformis embryos only showed swelling, except for one embryo which only 

developed one rooted shoot (Figure 5.2). 

R. flava embryos produced significantly more shoots when placed on a medium 

supplemented with either 2.3 µM kinetin or 4.7 µM kinetin compared to the control 

(Figure 5.3). The number of R. flava shoots observed on a medium with 4.7 µM 

kinetin was significantly higher than the number of shoots produced by all other 

treatments except on a medium supplemented with 2.3 µM kinetin. Single non-rooted 

shoots and swellings were observed on a medium with no plant growth regulators. 

Non-rooted shoots, rooted shoot clusters and swelling were observed on a medium 

supplemented with 2.3 µM kinetin. Only one of the shoot clusters on this medium did 

not produce a root, this shoot cluster had callus at its base. Non-rooted shoots, 

rooted and non-rooted shoot clusters and callus were produced on a medium 

supplemented with 4.7 µM kinetin. Shoot clusters and callus was observed on media 

supplemented with 23.2 µM kinetin. All but one of the shoot clusters formed on media 

supplemented with 23.2 µM kinetin did not root (Data not shown). Shoots generated 

from callus appeared vitrified and they also did not root on MS media with no plant 

growth regulators. 

The number of shoots of R. minutiflora and R. monadelpha produced on media 

supplemented with kinetin was much lower than the number of shoots produced by 

R. flava embryos and is not significantly different from the number of shoots 

produced on a medium with no plant growth regulators (Figures 5.4 and 5.5). Table 

4.5 in Chapter 4 shows that most of the embryos of these species show an in vitro 

growth response. In all cases where shoots were not formed, only swelling of the 

embryo was observed for these two species. One swollen embryo of R. minutiflora 

rooted on a medium with no plant growth regulators. Single shoots and one small 

122


shoot cluster (2 shoots) were formed on a medium supplemented with 2.3 µM kinetin. 

All shoots produced on this medium, except for one single shoot, were rooted. Only 

non-rooted single shoots developed from R. minutiflora embryos on a medium 

supplemented with 4.7 or 23.2 µM kinetin. For R. monadelpha embryos, only single 

rooted shoots were formed for all tested medium treatments. 

123


Figure 5.2: Effect of kinetin concentration on shoot production of Romulea diversiformis 

embryos after 2 months. Error bars indicate standard error of the mean. 

Figure 5.3: Effect of kinetin concentration on shoot production of Romulea flava embryos after 

2 months. Error bars indicate standard error of the mean. Different letters indicates 

significance differences between treatments according to Duncan’s multiple range test. 

124


Figure 5.4: Effect of kinetin concentration on shoot production of Romulea minutiflora embryos 

after 2 months. Error bars indicate standard error of the mean. 

Figure 5.5: Effect of kinetin concentration on shoot production of Romulea monadelpha 

embryos after 2 months. Error bars indicate standard error of the mean. 

125


No embryos of R. camerooniana or R. rosea showed any in vitro response. Embryos 

were completely dehydrated after 2 months. 

Because of the larger amount of replicates used for the vulnerable and attractive R. 

sabulosa it was possible to examine the specific in vitro response of embryos to 

different plant growth regulator combinations with some statistical confidence (Table 

5.1). The specific in vitro response for each treatment is therefore described in more 

detail for R. sabulosa than for other species. 

As with embryos of other species of R. sabulosa, some cultures showed only swelling 

(Table 5.1). A significantly higher percentage of cultures showed swelling on the 

control medium than for other treatments and roots were also produced on this 

medium. Shoot initiation was only observed in cultures with kinetin only. Although 

swelling with shoot initials was produced on all media with kinetin only, this response 

was significantly higher for the 2.3 M kinetin treatment compared to the control and 

all other treatments. Shoot and root formation was observed at the two lower kinetin 

concentrations. Shoot clusters were observed for the two higher kinetin 

concentrations. Significantly more shoot clusters were produced on a medium 

supplemented with 23.2 M kinetin compared to all other treatments, as 36% of the 

embryos that showed a growth response on this media produced shoot clusters. 

When the data was pooled it also showed that a medium supplemented with 23.2 M 

kinetin was best for shoot initiation. Among the embryos placed on media 

supplemented with 23.2 M kinetin that showed a growth response, 89% showed a 

shoot initiation response. Callus with shoot cluster initials was observed for embryos 

placed on media supplemented with 23.2 M kinetin. As with other species in 

Romulea, the shoots developed from these initials were however highly vitrified and 

did not develop roots when placed on a medium with no plant growth regulators. 

Cultures placed on five of the media treatments containing both kinetin and 2,4-D 

produced small (


observed when it was placed on media with activated charcoal. These cultures did 

not multiply well, or at all, and many became necrotic. 

What appeared to be indirectly formed embryogenic tissue was observed on all 

media supplemented with both kinetin and 2,4-D and its induction percentage was 

highest on a medium supplemented with 2.3 M kinetin and 2.3 M 2,4-D (Table 

5.1). Cultures that appeared to have direct embryogenesis was observed on media 

supplemented with 4.7 M kinetin and 2.3 M 2,4-D, and 23.2 M kinetin, 4.5 M 

2,4-D. Most of the cultures that appeared to have initiated embryogenesis were 

observed on the latter medium. 

The results of experiments that tested the effect of media supplemented with 5.4 M 

NAA and 2.3 M kinetin and media with two different concentrations of BA is not 

included. This is because of the vitrification observed in cultures placed on this media 

and the lack of subsequent explant response by these cultures when placed on 

media with no plant growth regulators. 

127


Table 5.1: Effect of kinetin and 2,4-D on excised embryos of Romulea sabulosa. Mean values in a column followed by different letters that indicates 

significance differences between treatments according to Duncan’s multiple range test (P 0.05). S = swelling of embryo; SR = swelling of embryo 

with rooting; SSI = swelling of embryo with shoot initials; SRF = shoot and root formation; SC = shoot cluster; SCR = shoot cluster with roots; 

CSCI = callus with shoot cluster initials; CIS = corm-like structure (< 10 mm); CAI = callus appearing incompetent; CPE = callus with potential 

embryogenesis; PDE = potential direct embryogenesis; CSGR = cultures showed growth response. Potential embryogenesis refers to cultures that 

appeared to develop embryo-like structures (Figure 5.7). 

Treatment ( M) 

Observed response (%) y 

S SR SSI SRF SC SCR CSCI CIS CAI CPE PDE CSGR 

control (no PGR’s) 43 ± 6 a 20 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 e 0 ± 0 b 53 ± 8 abc 

kinetin (2.3) 24 ± 3 bc 0 ± 0 b 24 ± 3 a 20 ± 0 a 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 e 0 ± 0 b 50 ± 10 abc 

kinetin (4.7) 30 ± 4 b 40 ± 0 a 20 ± 0 b 20 ± 0 a 20 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 e 0 ± 0 b 50 ± 7 abc 

kinetin (23.2) 20 ± 0 cd 0 ± 0 b 20 ± 0 b 0 ± 0 b 36 ± 8 a 20 ± 0 a 20 ± 0 a 0 ± 0 b 0 ± 0 b 0 ± 0 e 0 ± 0 b 63 ± 8 ab 

2,4-D (2.3) 20 ± 0 cd 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 20 ± 0 a 20 ± 0 de 0 ± 0 b 27 ± 7 c 

kinetin (2.3) + 2,4-D (2.3) 20 ± 0 cd 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 40 ± 7 ab 0 ± 0 b 47 ± 7 abc 

kinetin (4.7) + 2,4-D (2.3) 20 ± 0 cd 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 30 ± 6 a 0 ± 0 b 43 ± 6 a 27 ± 5 a 70 ± 7 a 

kinetin (23.2) + 2,4-D (2.3) 20 ± 0 bcd 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 20 ± 0 ab 0 ± 0 b 30 ± 5 abc 0 ± 0 b 50 ± 5 abc 

2,4-D (4.5) 30 ± 4 bc 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 20 ± 0 ab 20 ± 0 b 26 ± 5 cde 0 ± 0 b 37 ± 13 bc 

kinetin (2.3) + 2,4-D (4.5) 20 ± 0 cd 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 20 ± 0 ab 20 ± 0 a 35 ± 8 bcd 0 ± 0 b 47 ± 13 abc 

kinetin (4.7) + 2,4-D (4.5) 20 ± 0 cd 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 20 ± 0 ab 0 ± 0 b 33 ± 7 ab 0 ± 0 b 47 ± 7 abc 

kinetin (23.2) + 2,4-D (4.5) 20 ± 0 d 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 0 ± 0 b 20 ± 0 ab 20 ± 0 a 27 ± 5 cde 32 ± 5 a 57 ± 13 ab 

128


Significantly more shoots were produced on a medium supplemented with 23.2 M 

kinetin than a medium supplemented with 2.3 or 4.7 µM kinetin (Figure 5.6). Tukey’s 

HSD test revealed that the number of shoots formed on a medium supplemented 

with 23.2 M kinetin were significantly higher than on a medium supplemented with 

2.3 M kinetin (N = 14, 18; Mean difference = -3.42857; SE = 0.97832, p = 0.003). It 

also revealed that the number of shoots generated on a medium supplemented with 

23.2 M kinetin was significantly higher than on 4.7 M kinetin (N = 16, 18; Mean 

difference = -3.12500; SE = 0.94330, p = 0.005). 

Figure 5.6: Effect of kinetin concentration on shoot production of Romulea sabulosa after 2 

months. Error bars indicate standard error of the mean. Letters shows significance differences 

between treatments according to Tukey’s HSD test. 

Although it appeared that some cultures were embryogenic (Figure 5.7), sections of 

tissue presumed to exhibit direct and indirect embryogenesis did unfortunately not 

reveal any embryo initials and it was concluded that these cultures were not 

embryogenic, but formed abnormal shoot-like structures lacking chlorophyll. 

129


Figure 5.7: Visual observations of Romulea sabulosa cultures. Cultures including both kinetin 

and 2,4-D appears to exhibit embryo-like structures. 

5.3.3 Explant comparison 

Despite the apparent larger number of shoots produced between cultures in which 

embryos were used as explants compared to seedling organs, there is no significant 

difference between the numbers of shoots produced according to a paired t-test at 

95% confidence limits. There was also no significant difference between the number 

of shoots formed on media supplemented with kinetin and the number of shoots 

formed on media supplemented with other plant growth regulator treatments (Figure 

5.8). These results show that a medium supplemented with 23.2 µM mTR and 0.5 µM 

NAA significantly increases the amount of shoots formed per embryo compared to 

the control. This is not the case with seedling hypocotyls. Here cytokinins did not 

increase the number of shoots formed per explant (Figure 5.8). The number of 

shoots produced on a medium with 23.2 µM mTR and 0.5 µM NAA was not 

significantly different from the number of shoots observed on a medium 

supplemented with 2.3 µM kinetin or 2.3 µM mTR and 0.5 µM NAA. 

Explant response data is not shown, as there were no significant differences in 

explant response between the two explant types. Although these differences are not 

130


significant, they are important in comparing the two explant types and are described 

in the next few paragraphs. 

The percentage cultures that showed some growth response was higher for 

hypocotyls (92.3%) than embryos (67.3%). Embryos showed a higher percentage 

growth response on medium supplemented with both cytokinins and auxins (78.3%) 

than on a medium with only cytokinins (60.0%). This effect was not as pronounced 

for seedling hypocotyls; although the highest percentage response was observed for 

media with both cytokinins and auxins, there was only a 5% difference between the 

percentage response on these media and the percentage response on media with 

only cytokinins. 

The seedling hypocotyls showed higher rooting (53.8%) than the excised embryos 

(22.2%) on the control medium. A higher percentage of the shoots produced by 

embryos (33.3%) on a medium supplemented with 0.5 µM kinetin were rooted 

compared to seedling hypocotyl cultures (15.8%) on the same medium. On a 

medium supplemented with 2.3 µM kinetin, seedling hypocotyls produced a higher 

amount of rooted shoots (26.3% compared to 11.8% for embryos). None of the 

shoots produced by excised embryos on a medium supplemented with 23.2 µM 

kinetin rooted, whereas 36.8% of the shoots produced by seedling hypocotyls on this 

medium rooted. None of the shoots produced by excised embryos on a medium 

supplemented with 2.3, 4.7 or 23.2 µM kinetin or 0.5 µM NAA rooted, whereas at 

least 20% of shoots produced by seedling hypocotyls on these media were rooted. 

The seedling hypocotyls showed higher rooting percentage (>30%) than the excised 

embryos (


For seedling hypocotyl and excised embryo cultures a large volume of callus was 

observed on more than 50% of cultures on media supplemented with cytokinins and 

auxins. For excised embryos, a small number (


There was a significant difference between the number of shoots produced on media 

supplemented with 2.5 M mTR and the number of shoots produced on media 

supplemented with all the kinetin concentrations tested including 2.5 M kinetin 

(Figure 5.9). Although the highest number of shoots were also produced on media 

supplemented with 2.5 M mTR, there was no significant difference between the 

number of shoots produced on this medium and the number of shoots produced on a 

medium supplemented with 2.5, 5.0 or 7.5 M BA, 5.0 or 7.5 M mTR, or 2.5 or 5.0 

MemTR. 

The shoots produced on a medium supplemented with 2.5 M mTR appeared 

healthy and 40% of them were rooted. All other media treatments also produced 

healthy shoots of which at least 20% were rooted, except for media supplemented 

with BA, on which vitrified shoots were observed as with previous experiments. 

Average number of shoots per explant 

8 

6 

4 

2 

0 

cd 

Control 

d 

2.5 µM kinetin 

cd 


bcd 


ab 

2.5 µM BA 

abcd 

5.0 µM BA 

abc 

7.5 µM BA 

bcd 

2.5 µM mT 

bcd 

5.0 µM mT 

cd 

7.5 µM mT 

a 

2.5 µM mTR 

Media treatments 

abcd 

5.0 µM mTR 

abcd abc 

Figure 5.9: Effect of three different concentrations of five cytokinins on multiplication of 

Romulea sabulosa shoots after 2 months. Error bars indicate standard error of the mean. 

Letters shows significant differences between treatments according to Duncan’s multiple 

range test. 

7.5 µM mTR 

2.5 µM MemTR 

abcd 

5.0 µM MemTR 

bcd 

7.5 µM MemTR 

133



Preliminary experiments with R. diversiformis, R. flava, R. leipoldtii and R. minutiflora 

seedlings showed that hypocotyls were the only seedling organs that gave an in vitro 

response for the Romulea species tested. They also show that R. leipoldtii and R. 

flava form the highest number of shoots per seedling hypocotyl compared to the 

other species tested. Because of possible vitrification and culture incompetence, the 

use of BA and shoots generated from callus should be avoided for this genus. 

The in vitro response of R. diversiformis embryos and hypocotyls, although 

morphologically distinct, is equally unproductive, as no shoots were formed after 2 

months. A higher concentration of cytokinins, or different cytokinins, is perhaps 

needed to generate a more pronounced shooting response for explants of this 

species. 

The rooting of shoots formed in vitro of all species, except R. diversiformis and R. 

monadelpha was inhibited to a certain degree by a medium supplemented with 23.2 

M kinetin. Experiments performed with the aim of finding a concentration of kinetin 

high enough to induce shooting, but low enough to allow for normal root growth, is 

necessary for the efficient production of healthy shoots of these species. Such an 

experiment should investigate the effect of various kinetin concentrations between 

7.5 and 23.2 µM. 

Although R. leipoldtii embryos produced significantly more shoots on a medium with 

23.2 µM mTR and 0.5 µM NAA, this medium and a medium supplemented with 2.3 

µM mTR and 0.5 µM NAA resulted in a higher percentage of cultures with callus. This 

indicates that although these are the best media for culture initiation, they are not 

suitable for multiplication of shoots of R. leipoldtii. Shoots produced from this callus 

were vitrified. 

The fact that there is no significant difference between the number of shoots 

produced by embryos and hypocotyls of R. leipoldtii suggests that excising the 

embryos of even the smallest seeds of all species presented here, except for R. 

134


camerooniana and R. rosea (no in vitro response was observed for these species), 

can be used as a faster method for obtaining shoot cultures. The germination of the 

embryos of these species does not require the low temperature and stratification 

treatments that the seeds need as germination cues and the embryo geminates 

within about 2 weeks, whereas seeds require at least 2 months. 

Although a higher percentage of R. leipoldtii hypocotyls than embryos showed some 

growth response because of their larger size, embryos were more responsive to the 

plant growth regulators in the medium because of the younger physiological age of 

these tissues (SMITH, 2000). 

Although the rooting of embryo-derived shoots of R. leipoldtii was inhibited by high 

concentrations of cytokinins and developed more callus compared to hypocotyl- 

derived shoots it should be considered that the hypocotyls are essentially fully 

developed plants with trimmed shoots and roots, so that the development of an entire 

new shoot is not necessary. Following this logic, it also means the single shoots 

observed in some cultures for which hypocotyls is the initial explant was in fact the 

original hypocotyl, which did not form more shoots but only elongated and, in most 

cases, produced roots. When the data is viewed from this perspective, embryos 

produced more shoots than hypocotyls. 

These factors make the choice of explant type very difficult. The explant type used 

should therefore rather depend on the species used. 

These results show that the species response in Romulea is influenced to great 

extent by genetic factors, as media treatments or explant type could not increase the 

shooting of R. diversiformis, R. minutiflora and R. monadelpha to numbers similar to 

that of R. flava and R. sabulosa and the rooting of some species is not suppressed 

by a medium with a high kinetin concentration while the rooting of other species is 

inhibited on such a medium. 

Although the kinetin concentrations tested in the R. sabulosa shoot multiplication 

experiment was not the same as those used in previous experiments, the number of 

shoots produced after two months on a medium supplemented with 2.5 µM mTR (5.5 

135


± 1.3 SE) is much higher than the number of shoots produced by a medium 

supplemented with 23.2 µM kinetin (2.4 ± 0.6 SE) in previous experiments and the 

shoots appeared healthier. This medium is therefore better suited for R. sabulosa 

shoot multiplication. The effect of similar concentrations of mTR on shoot culture 

initiation from embryos of R. sabulosa should be tested. 

It was expected that the direct shoot organogenesis requirements for Romulea 

species would be similar to that of Crocus species due to the small phylogenetic 

distance between these two genera compared to other genera in Iridaceae that has 

been micropropagated (REEVES et al., 2001). The abnormal growth observed on 

media supplemented with BA and 2,4-D for Romulea cultures however shows that 

this is not the case. Most studies on direct shoot organogenesis in Iridaceae species, 

summarized in Table 2.6 in Chapter 2, also show BA and 2,4-D to be suitable plant 

growth regulators for direct shoot organogenesis. Kinetin has however been a 

component in a media for direct shoot organogenesis, in the absence of BA and 2,4- 

D, for a number of Gladiolus species (ZIV et al., 1970; LILIEN-KIPNIS & KOCHBA, 

1987; ZIV & LILIEN-KIPNIS, 2000). In a study on Sisyrinchium laxum and Tritonia 

gladiolaris, shoots generated on media supplemented with mT appeared healthier 

than those generated from media supplemented with BA, which had abnormal and 

stunted growth, and no roots (ASCOUGH et al., 2011). This is analogous to positive 

effects of a topolin and the negative effect of BA observed in this study. 

5.5 SUMMARY 

• Before these experiments were conducted there were no studies published on 

the micropropagation of Romulea species. 

• Both embryos and seedling hypocotyls can be used for R. flava, R. leipoldtii 

and R. minutiflora in vitro shoot culture initiation 

• R. sabulosa shoot cultures can only be initiated by using embryos as explants, 

because the lack of germination for this species. 

• Shoot cultures of R. diversiformis, R. camerooniana and R. rosea could not be 

initiated due to the lack of an in vitro explant shooting response. 

136


• Shoot cultures was initiated on media supplemented with 2.3 to 23.2 M 

kinetin for R. flava, R. leipoldtii, R. minutiflora, R. monadelpha and R. 

sabulosa, with the most suitable concentration depending on the species 

used. 

• A medium supplemented with 2.5 M mTR is suitable for R. sabulosa shoot 

multiplication. 

• BA caused vitrification of shoots in all the experiments in which it was used 

and is not a suitable cytokinin for the micropropagation of these species. 

137

6 In vitro corm formation and flowering and ex vitro 

acclimatization 


In vitro formation of storage organs increases the ex vitro survival rate during 

acclimatization and these organs serve as a more attractive product than seeds, as 

flowering of the plant can be enjoyed at a much earlier stage (ASCOUGH et al., 

2009). Commercialising in vitro produced corms of R. sabulosa will also reduce the 

pressures on the vulnerable populations from which seeds are harvested 

(RAIMONDO et al., 2009). 

In vitro flowering is an important tool in ornamental plant breeding, as it enables the 

breeder to see the floral traits in a much shorter time than in ex vitro conditions. 

Some factors that influence in vitro flowering and ex vitro survival is discussed in 

section 2.10 and 2.8.9 of Chapter 2 respectively. 

The aims of this Chapter were to establish an in vitro corm induction protocol for R. 

minutiflora, R. leipoldtii and R. sabulosa, to investigate the effect of some physical 

and chemical stimuli on in vitro flowering of R. minutiflora and R. sabulosa corms and 

to establish an ex vitro acclimatization protocol for R. minutiflora and R. sabulosa 

corms and plantlets. 


6.2.1 Corm formation 

In all experiments, the in vitro generated shoots were separated from each other, 

trimmed to 25 mm and all roots were removed for uniformity. If not stated otherwise a 

MS medium supplemented with 100 mg.l -1 myo-inositol and 3% sucrose, with pH 

adjusted to 5.7 and solidified with 0.8% agar was used. All experiments were 

conducted in a laminar flow hood and cultures were placed in a growth chamber 

under 3.4 µmol m –2 s –1 light using Osram ® 75 W cool white fluorescent tubes with a 

16/8 light/dark photoperiod. The duration of all experiments was 6 months, as no 

138

In vitro corm formation and flowering and ex vitro acclimatization 

corm formation was observed after 2 and 4 months and no tunic development was 

seen after 5 months. 

The shoots of R. minutiflora produced from seedling hypocotyls on a medium 

supplemented with 23.2 µM kinetin and 5.4 µM NAA were multiplied on the same 

medium for 6 months. These shoots were placed singly in 33 ml culture tubes on 10 

ml of MS media supplemented with 3%, 6% or 9% sucrose or 5.0 g.l -1 activated 

charcoal. The tubes were then placed in growth chambers maintained at 10°C, 15°C, 

20°C, 25°C or 30°C. 

The remainder of these shoots were multiplied for a further 4 months and 

subsequently used in an experiment to test the effect of growth retardants on corm 

formation in R. minutiflora. The shoots were placed in 33 ml culture tubes on 10 ml of 

MS medium supplemented with 3.4, 17.0 or 34.0 M paclobutrazol (PP3), or 4.9 M 

abscisic acid (ABA). These tubes were placed at 25°C under 4.3 µmol m –2 s –1 light 

supplied by Osram ® 75 W cool white fluorescent tubes. Twenty replicates were used 

per treatment. 

The shoots produced by seedling hypocotyls of R. leipoldtii on a medium 

supplemented with 22.2 µM BA were multiplied on this medium for 4 months. The 

shoots were placed in 33 ml culture tubes on 10 ml of MS medium supplemented 

with 3%, 6% or 9% sucrose or 5.0 g.l -1 activated charcoal. These tubes were placed 

at 25°C under 4.3 µmol m –2 s –1 light supplied by Osram ® 75 W cool white fluorescent 

tubes and 50 replicates were used per treatment. 

The shoots of R. sabulosa produced from excised embryos on a medium 

supplemented with 23.2 µM kinetin were multiplied on the same medium for 6 

months. These shoots were placed singly in 33 ml culture tubes on 10 ml of MS 

media supplemented with 3%, 6% or 9% sucrose, 5.0 g.l -1 activated charcoal or 23.2 

µM kinetin. The tubes were then placed in growth chambers were maintained at 

10°C, 15°C, 20°C, 25°C or 30°C. Ten shoots were used per treatment and the 

experiment was repeated three times. 

Corm induction percentage and corm diameter for different temperatures and media 

compositions were analyzed for significant differences with Duncan’s multiple range 

139


test using Genstat. To examine the combined effect of treatments on corm induction 

rate and corm weight, product analysis was done. This was done by multiplying the 

proportion of corm induction with the mass of the corms produced (mg) and dividing 

this number with hundred according to the methods of ASCOUGH et al. (2008). A 

high value indicates that the treatment resulted in a high proportion of induction and 

corm mass, while a low value indicates that either the proportion of induction, the 

corm mass or both are low. 

6.2.2 In vitro flowering 

After an in vitro flower was observed in the R. minutiflora corm formation treatments 

(9% sucrose at 20°C) further experiments to repeat this result were initiated. 

Information from UYEMURA & IMANISHI (1984), FLAISHMAN & KAMENETSKY 

(2006), LIGHT et al., (2007) and a study by KÖK (2007) on the difference in soil 

composition during vegetative growth and flowering in R. columnae was used to 

design an in vitro flowering experiment. 

As it is better to use larger storage organs for florogenesis studies, only the largest 

corms were selected (FLAISHMAN & KAMENETSKY, 2006). These corms were 

placed in test tubes in a growth chamber with high light (190.1 µmol m –2 s –1 ) set at 

20°C. Medium treatments included a half strength MS medium, a half strength MS 

medium with 9% sucrose, a full strength MS medium with 9% sucrose, a full strength 

MS medium with 1:500 (v/v) smoke water and a full strength MS medium with 3% 

sucrose as a control. For each treatment 20 corms of R. minutiflora and R. sabulosa 

between 5 and 10 mm in diameter were used. 

6.2.3 Ex vitro acclimatization and corm viability 

Before all corms and plantlets were used for ex vitro growth studies they were rinsed 

in autoclaved distilled water to remove agar from roots. Corms were partially dried in 

vitro by placing them inside a Petri dish on a laminar flow bench for two weeks before 

planting. 

The mist-house and greenhouse used in these experiments is situated in the 

University of KwaZulu Natal Botanical Garden in Pietermaritzburg (30° 24’ E, 29° 37’ 

S, 655 m above sea level). The mist-house had a misting interval of 15 min and a 

misting duration of 10 s, whereas the greenhouse did not have an automated 

140


watering regime at the time during which these experiments where conducted and 

plants were watered every second day with 500 ml of water. 

Corms of R. minutiflora and R. sabulosa were planted in plastic planting trays with a 

1:1 ratio of potting soil and sand. For each species, 6 corms were placed in 6 trays. 

The trays were placed in the mist-house for 24 h, after which they were moved to the 

greenhouse. The trays were only placed in the mist-house for this short period of 

time because all corms were necrotic after only two weeks in an initial experiment 

were 2 trays with 6 corms each where placed in the mist-house, suggesting that 

these conditions are too moist. 

In another experiment corms of R. sabulosa were planted in clay pots with a 1:1 ratio 

of potting soil and sand. Because of the limited number of pots, 4 corms were placed 

in each pot and 30 pots were used. These pots were placed in a growth chamber 

with high light (190.1 µmol m –2 s –1 ) maintained at 20°C. 

Healthy rooted plantlets of R. minutiflora and R. sabulosa that did not produce corms 

during corm formation experiments were planted in plastic planting trays with a 1:1 

ratio of potting soil and sand. For each species, 6 plantlets were placed in 5 trays. 

The trays were placed in the mist-house for 24 h, after which they were moved to the 

greenhouse. 

A small experiment was conducted with the remaining corms formed in vitro at 10 °C. 

Four corms were placed in a clean plastic container with autoclaved vermiculite 

moistened with 50% Hoagland’s nutrient solution. The containers were placed in a 

growth chamber set at 20°C with a 16 h photoperiod of 12.5 µmol m –2 s –1 irradiance 

for two months. The rest of the corms were used for a viability study. The viability of 

ten corms larger than 5 mm in diameter were tested according to the method of 

WAGNER (1984). 

141

6.3 RESULTS 

6.3.1 Corm formation 


In the first corm formation experiments with shoots of R. minutiflora, no corms were 

observed at temperatures of 25°C and higher. The results of 15°C are not reported 

as the growth chamber malfunctioned and the cultures had to be discarded. No 

significant differences in corm induction from shoot explants were observed either 

when changing the temperature or altering the sucrose concentration (Table 6.1). 

Corm mass, however, increased with increasing sucrose concentration. This was true 

at both 10 and 20 °C. The addition of activated charcoal to media with 3% sucrose 

significantly increased corm mass under both temperature regimes (Table 6.1). The 

product of corm induction and corm mass indicates that placing the shoots at 10°C 

on a medium with 6% sucrose leads to the best combination of both proportion of 

induction and corm mass, followed by the shoots placed at 20°C on media with 9% 

and 3% sucrose respectively (Table 6.1). 

Table 6.1: The effect of different temperatures and media composition on the in vitro formation 

and growth of Romulea minutiflora corms. Data shows the means ± the standard error. Letters 

indicates significant differences between treatments according to Duncan’s multiple range 

test. 

Culture temperature and medium 

treatment 

Corm induction % Corm mass (mg) Product 

10°C 3% sucrose 65.2 ± 9.2 a 108.7±21.9 b 70.9 

6% sucrose 80.4±6.9 a 279.0±114.0 ab 224.3 

9% sucrose 81.9±9.1 a 134.5±19.6 b 110.2 

5.0 g.l -1 activated charcoal 56.7±5.6 a 285.2±33.7 a 161.7 

20°C 3% sucrose 69.4±4.6 a 152.0±31.3 b 197.9 

6% sucrose 73.9±3.9 a 227.0±61.1 b 105.5 

9% sucrose 72.5±10.3 a 273.7±50.3 a 198.4 

5.0 g.l -1 activated charcoal 63.8±5.9 a 239.3±45.0 a 157.7 

PP3 and ABA both stimulated corm formation at 25°C in the second experiment with 

R. minutiflora. The addition of 17.0 and 34.0 M PP3 resulted in the largest 

percentage corm induction and corm size (Table 6.2). 

142


Table 6.2: Percentage corm induction for Romulea minutiflora shoots cultured on medium 

supplemented with growth retardants. 

Treatment Basal swelling 

(%) 

Corm induction 

(%) 

Corms > 5 mm 

(%) 

Control 70% 0% 0% 

3.4 M PP3 55% 15% 10% 

17.0 M PP3 50% 35% 20% 

34.0 M PP3 35% 35% 20% 

4.9 M ABA 45% 20% 5% 

No corm formation was observed for any of the temperature and medium treatments 

tested for R. leipoldtii. When 100 bottles of shoots multiplied on media supplemented 

with BA placed at 25°C were not subcultured for 6 months no corm formation or 

basal thickening was observed. This was not the case with the shoot cultures placed 

at 25°C multiplied with media supplemented with kinetin and topolins. Here corms or 

basal thickening was observed for all shoots not multiplied after 6 months. 

Such basal thickening was observed to a certain extent for non-subcultured shoot 

cultures of R. minutiflora and R. sabulosa, but no corm formation was observed in 

any cultures placed at 25°C for these species. 

The percentage corm induction for R. sabulosa after 6 months was significantly 

higher at 10 and 20 °C with all tested sucrose concentrations compared to 15 °C 

(Table 6.3). Corm induction was inhibited at 25 °C in this experiment (data not 

shown). The largest corm mass was recorded for shoots placed at 15 °C on a MS 

medium supplemented with 6% sucrose (Table 6.3). The mass of these corms were 

however not significantly different from the mass of the corms obtained for cultures 

placed at the same temperature on a medium with 9% sucrose or the mass of corms 

produced on a medium with activated charcoal and the mass of corms at 20 °C 

placed on a medium with 6% sucrose. Product analysis shows that corms at 15 °C 

with 6% sucrose achieved the highest value (Table 6.3). 

143


Table 6.3: The effect of different temperatures and media composition on the in vitro formation 

and growth of Romulea sabulosa corms. Data shows the means ± the standard error. Letters 

indicate significant differences between treatments according to Duncan’s multiple range test. 

Culture temperature and medium Corm induction Corm mass (mg) Product 

treatment 

(%) 

10°C 3% sucrose 100.0 ± 0.0 a 325.5 ± 39.0 b 325.5 

6% sucrose 100.0 ± 0.0 a 346.5 ± 56.3 b 346.5 

9% sucrose 100.0 ± 0.0 a 294.1 ± 49.6 b 294.1 

5.0 g.l -1 activated charcoal 87.5 ± 7.2 abc 336.3 ± 50.8 b 294.2 

23.2 µM kinetin 100.0 ± 0.0 a 371.1 ± 49.8 b 371.1 

15°C 3% sucrose 71.3 ± 10.9 bc 325.2 ± 21.9 b 231.7 

6% sucrose 81.7 ± 6.9 bc 522.7 ± 61.7 a 426.9 

9% sucrose 

5.0 g.l 

83.8 ± 5.5 bc 439.3 ± 71.4 ab 348.9 

-1 activated charcoal 68.8 ± 12.6 c 409 ± 59.4 ab 281.2 

23.2 µM kinetin 20.0 ± 14.1 d 295.3 ± 39.2 b 58.5 

20°C 3% sucrose 100.0 ± 0.0 a 310.9 ± 54.7 b 310.9 

6% sucrose 100.0 ± 0.0 a 383.5 ± 36.6 ab 383.5 

9% sucrose 

5.0 g.l 

100.0 ± 0.0 a 372.7 ± 57.8 b 372.7 

-1 activated charcoal 91.7 ± 8.3 ab 362.6 ± 63.8 b 332.4 

23.2 µM kinetin 18.8 ± 18.8 d 281.8 ± 52.9 b 52.2 

Multiple corm formation from the same shoot was observed in some cultures of R. 

sabulosa (Table 6.4). The highest percentage of multiple corm formation was 

observed at 15°C on a medium supplemented with 23.2 µM kinetin, while the highest 

number of corms per shoot was observed at 10°C on a medium with 3% sucrose. 

Multiple corm formation also occurs in Romulea species under natural environmental 

conditions (DE VOS, 1972). 

Table 6.4: Cultures with multiple corm formation for Romulea sabulosa. This shows the 

percentage of corm formation in cultures in which corm formation observed (Total cultures 

with corms) and the average number of corms produced in instances of multiple corm 

formation. 

Culture temperature and medium 

treatment 

Multiple corm 

formation (%) 

Total cultures 

with corms 

Average number of 

multiple corms 

10°C 3% sucrose 6.3% 16 7.0 

9% sucrose 16.7% 18 2.3 

5.0 g.l -1 activated charcoal 6.3% 16 3.0 

15°C 6% sucrose 7.1% 14 3.0 

9% sucrose 12.5% 16 2.0 

5.0 g.l -1 activated charcoal 7.7% 13 2.0 

23.2 µM kinetin 50.0% 4 2.0 

20°C 3% sucrose 9.1% 11 2.0 

144

6.3.2 In vitro flowering 


Although an in vitro formed flower was observed for R. minutiflora during corm 

formation experiments at 20°C on a medium with 9% sucrose (Figure 6.1), none of 

the further experiments yielded any flowers or flower initials. 

Figure 6. 1: An in vitro formed flower of Romulea minutiflora observed in a test tube placed at 

20°C on a medium with 9% sucrose. 

6.3.3 Ex vitro acclimatization and corm viability 

None of the corms or plantlets survived more than one month. Four corms planted 

inside the plastic container however survived (Fig. 6.2) for two months, and 7 out of 

10 corms tested were viable. 

145


Figure 6.2: Corms of Romulea sabulosa growing in a modified plastic container with 

vermiculite after 2 months. Bar = 20 mm. 

146



Temperature is the major factor that affects storage organ morphogenesis 

(ASCOUGH et al., 2009). Low temperature significantly increased corm formation in 

R. minutiflora and R. sabulosa. The highest number of instances of multiple corm 

formation for R. sabulosa was also observed at 10 and 15°C. 

In a study by ASCOUGH et al. (2008) on Watsonia vanderspuyiae (in the same 

subfamily, Ixioideae) the highest percentage corm induction was observed for shoots 

placed at low temperatures. This study showed no significant difference between 

corm induction percentages of shoots placed on MS media supplemented with 

sucrose at 10 and 20 °C. Corm formation at low temperatures has been observed in 

vitro for many genera in the Iridaceae (ASCOUGH et al., 2009). An exception to this 

is Crinum macowanii, where storage organ formation occurs above 25 °C and is 

inhibited at lower temperatures (SLABBERT et al., 1993). The fact that corm 

induction was inhibited for R. minutiflora and R. sabulosa at 25 °C in this study 

correlates with the results of Gladiolus spp., where corm induction was inhibited at 

this temperature (TAN NHUT et al., 2004). 

In this study the requirements for corm formation of R. leipoldtii is similar to that 

determined in a study involving Crocus sativus corm formation (PLESSNER et al., 

1990). It was however shown by HOMES et al. (1987) that corm formation of Crocus 

sativus also occurred at 30°C, a temperature that totally inhibits the corm formation of 

R. minutiflora and R. sabulosa and that would probably inhibit the corm formation of 

R. leipoldtii as it occurs in the same regions as R. minutiflora (Figure 2.1). 

When PP3 and ABA were added to the medium on which R. minutiflora shoots were 

placed, these shoots developed corms although they were placed at 25°C, a 

temperature that totally inhibits corm formation when these retardants that reduce 

leaf elongation and promote storage organ formation are not present. This effect of 

PP3 has also been shown for Dierama and Gladiolus species (MADUBANYA, 2004; 

STEINITZ & LILIEN-KIPNIS, 1989; ZIV, 1989; ZIV et al., 1998). 

147


The fact that corms were not observed in any R. leipoldtii shoot cultures 

supplemented with BA after 6 months show that this chemical inhibits corm formation 

for this genus. The same effect is expected for the other species in this genus, as the 

shoots of all species generated and multiplied on BA appeared abnormal and their 

root growth was stunted. 

In Watsonia cultures, a correlation between corm mass and carbohydrate 

concentration was observed, with corm induction in some species decreasing as the 

carbohydrate concentration increases (ASCOUGH et al., 2008). In the present study 

this was also observed to some extent, as the treatment that delivered the highest 

corm mass had a 6% sucrose concentration. It was however surprising that elevated 

levels of carbohydrates did not have a statistically significant effect on corm mass. 

A two step corm formation protocol would work best for R. sabulosa, as the corms 

differentiate and accumulate carbohydrates under different temperatures. This two 

step system would involve placing corms at either 10 or 20°C for a few months and 

then transferring these cultures to 15°C. 

A similar two step program was proposed by ASCOUGH et al. (2011) for Tritonia 

gladiolaris. This two step system forms corms at lower temperatures (10 and 15°C) 

and promotes the accumulation of carbohydrates at higher temperatures (20°C). 

They suggest that the physiological mechanisms involved here are probably an 

adaptation for survival, so that the low temperature is perceived as a cue for corm 

induction and the onset of dormancy before the approach of unfavourable conditions. 

Corm production is however promoted at 10 and 20°C and corm mass increases at 

15°C, a temperature flanked by the two former temperatures, for R. sabulosa. This 

phenomenon may be explained by comparing the life cycle of R. sabulosa with the 

temperature, rainfall and humidity observations for the last five years in 

Nieuwoudtville. 

R. sabulosa flowers from August to September. Before flowering there are a few 

weeks of vegetative growth, facilitated by the increase in rainfall in July. During July 

and August the averaged daily minimum is 5.1 ± 0.3°C and 4.7 ± 0.2°C and the 

averaged daily maximum temperature is 18.6 ± 0.6°C and 18.1 ± 0.7°C respectively. 

148


During this period new axillary buds are formed. The axillary buds gradually form 

their own corm at the base as conditions for growth becomes less suitable at the end 

of flowering in September. During September and October the averaged daily 

minimum temperatures are 6.3 ± 0.7°C and 8.2 ± 0.3°C and the averaged daily 

maximum temperatures are 21.3 ± 0.7°C and 25.6 ± 0.4°C. The maximum 

temperature of periods of vegetative growth and corm induction is close to 15°C and 

the minimum and maximum temperatures are close to 10 and 20°C during periods of 

senescence, when accumulated carbohydrates may be reallocated to the corm in the 

natural habitat of R. sabulosa. Although such an explanation is plausible, the 

interactions of factors that influences the climate of Namaqualand is very complex 

and it is therefore a difficult topic (DESMET, 2007). 

The corm production rate of various Gladiolus cultivars was also increased by a two 

step bud-culture technique, involving short-term exposure to a medium with plant 

growth regulators and subsequent withdrawal from plant growth regulators in a liquid 

medium (SEN & SEN, 1995). 

The specific temperature needed for carbohydrate accumulation for R. sabulosa, 

compared to that of Tritonia gladiolaris may be explained by its very restricted 

distribution compared to that of the widespread T. gladiolaris. An adaptation to 

accumulate carbohydrates above a certain temperature and to induce corms below 

this temperature would allow a species to spread to a variety of habitats where this 

temperature is observed, whereas a specific carbohydrate accumulation temperature 

range will limit the populations to an area with periods of this temperatures long 

enough to accumulate sufficient carbohydrates to be able to survive the dry seasons 

underground in a dormant state and to produce enough vegetative growth in the next 

season for the accumulation of additional carbohydrates. 

The fact that corms did not flower in vitro is probably because these plants need 

much colder minimum temperatures (below 5°C) such as in their natural environment 

for growth and development. The one corm of R. minutiflora which flowered in vitro at 

20°C was perhaps a mutation. 

Plantlets of Romulea species develop viable corms after 6 months, which can be 

commercialized as propagation units. Corm formation speeds up the 

149


micropropagation and seedling establishment process and it is also cost-effective, as 

there is no need for hardening or acclimatization of plants. 

6.5 SUMMARY 

• Low temperature significantly increased corm formation in R. minutiflora and 

R. sabulosa 

• A two step corm formation protocol involving placing corms at either 10 or 

20°C for a few months and then transferring these cultures to 15°C should be 

used for R. sabulosa 

• When PP3 and ABA is added to the medium on which R. minutiflora shoots 

were placed, these shoots develop corms at 25°C, a temperature that totally 

inhibits corm formation when these growth retardants are not present 

• BA inhibited corm formation in R. leipoldtii 

• Corms of Romulea sabulosa formed at 10°C are viable 

• Corms can be commercialized as propagation units to grown in winter-rainfall 

areas with minimum temperatures below 5°C during winter 

150

7 Commercialization potential of Romulea species 

Attributes that make species of Romulea attractive are their beautiful flowers and growth 

forms (Figure 7.1). The flowers are not only attractive because of their wide range of 

colours, including yellow to white, pink, orange, apricot, red, magenta, lilac and purple, 

but also because of the interesting shape of some flowers such as R. diversiformis 

(Figure 7.1 C) and R. hantamnensis (DE VOS, 1972; MANNING & GOLDBLATT, 2001). 

The flowers of some species are also scented. These include the honey scented flowers 

of R. austinii and the honey and coconut scented flowers of R. schlechteri. The growth 

form of many Romulea species is very reduced, with subterranean stems and a few 

filiform leaves. This accentuates the beautiful flowers of these species and creates a 

spectacular floral display when flowers are grown together. 

Figure 7.1: Showing eight species used in propagation experiments arranged from the largest to 

the smallest growth form. From the left they are Romulea minutiflora (A), R. camerooniana (B), R. 

diversiformis (C), R. rosea (D), R. flava (E), R. leipoldtii (F), R. monadelpha (G) and R. sabulosa (H). 

Modified from DE VOS (1972) and photographs taken by Dr. John C. Manning. Horizontal bar = 50 

mm. 

151

Commercialization potential of Romulea species 

Many horticulturalists I have spoken to during this study are not just interested in 

Romulea because of their beautiful flowers, but also because growing seeds of Romulea 

to the flowering stage is considered somewhat of a horticultural feat and they aspire to 

the challenge. There is a demand for propagative material of these plants and many 

companies supplying seeds via the internet. These companies often supply inadequate 

information on propagation. The seed germination protocols reported in Chapter 4 can 

now be used by horticulturalists as more effective methods for germinating seeds of R. 

diversiformis, R. flava, R. leipoldtii, R. minutiflora, R. monadelpha and R. rosea. 

Germination of the seeds of some of the most attractive species in this genus is however 

low. 

In this study, micropropagation protocols that produce corms as an end-product have 

been established for R. leipoldtii, R. minutiflora and R. sabulosa. Corms of these species 

can now be commercialised. For the attractive R. sabulosa, one embryo produces 2.1 ± 

0.7 SE shoots after 2 months; placing these shoots on a medium supplemented with 2.5 

µM mTR for a further 2 months multiplies their number by 5.5 ± 1.3 SE. Each of these 

shoots can then be induced to produce a corm after 6 months. This means that 1 

embryo can produce about 12 corms after 10 months or about 65 corms after 12 months 

(if shoots are subcultured to medium supplemented with 2.5 µM mTR for another 2 

months). Although corms of R. flava and R. monadelpha were not produced, stems with 

basal thickening, like those observed for R. leipoldtii, R. minutiflora and R. sabulosa, 

were observed in shoot cultures not subcultured for 4 months. It is therefore expected 

that these species would also produce corms when placed at low temperatures. 

Some other attractive species that were not micropropagated in this study include R. 

citrina and R. tabularis, which were used in initial germination experiments, and R. 

eximia for which seeds could not be obtained. 

R. eximia flowers are old-rose pink to dark old-rose or deep red (DE VOS, 1972; 

MANNING & GOLDBLATT, 1996; MANNING & GOLDBLATT, 2001). The cup is 

greenish to pale yellow and is streaked purple. There are dark red blotches on each 

152


segment around the cup. R. eximia flowers from August to September (MANNING & 

GOLDBLATT, 2001). According to DE VOS (1972) this plant has 2 to 3 flowers or more. 

R. eximia is in the same subgenus and section as R. sabulosa, whereas R. citrina and 

R. tabularis are in the same subgenus, section and series as R. leipoldtii and R. 

minutiflora (MANNING & GOLDBLATT, 2001). R. austinii and R. schlechteri, species 

with scented flowers, are also in the same subgenus, section and series as R. leipoldtii 

and R. minutiflora. 

It is therefore very likely that the culture requirements for R. eximia could be very similar 

to that of R. sabulosa and that the culture requirements for R. citrina, R. tabularis, R. 

austinii and R. schlechteri could be very similar to that of R. leipoldtii and R. minutiflora. 

This study can therefore be useful for the commercialization of these species. 

Apart from the increased multiplication rate in vitro compared to conventional 

propagation, in vitro techniques such as embryo rescue established for these species 

will also be useful in the commercialization of these species. Embryo rescue enables the 

development of the embryo in vitro if the endosperm is underdeveloped (HARTMANN & 

KESTER, 1965). Crosses have been attempted in vivo on numerous species of 

Romulea by DE VOS (1972). She found that crosses between species that were not in 

the same section and with different chromosome numbers were largely unsuccessful. 

When crossing species in different subsections and sections she found that seeds were 

not viable. She speculated that this was because of a failure of endosperm 

development, suggesting that the embryo may be viable and that the techniques of 

embryo rescue and in vitro culture could be used to cultivate these hybrids (DE VOS, 

1972). Weak plants were obtained in a few cases. These plants died within the first 

growth season, before flowering (DE VOS, 1972). 

Using embryo rescue techniques, the beautiful and large flowered species in the 

subsection Spatalanthus (e.g. R. sabulosa and R. monadelpha) can therefore 

theoretically be crossed with species with scented flowers such as R. austinii and R. 

schlechteri in the subsection Romulea. Such a cross may result in a phenotype with 

153


large, beautiful and scented flowers which can be cloned in vitro using protocols 

established in this study. Embryo rescue and in vitro culture techniques can also be 

used to cross such a phenotype with a widespread species such as R. rosea, so that 

cultivation requirements will be less specific. The commercialization potential of such a 

phenotype would be very high, as it could be cultivated in numerous geographical 

locations by somebody that does not have extensive horticultural experience. 

Natural hybridization amongst the South African Romulea species are rare (DE VOS, 

1972). Intermediates between R. rosea var. australis and var. communis have been 

found. A similar case of natural hybridisation has been reported for R. rosea var. reflexa 

and var. australis. DE VOS (1972) reported similar cases of natural hybridization of 

variants within species for R. cruciata and R. obscura. She also reported a case of 

natural hybridisation of R. leipoldtii and R. tabularis and speculated that R. sabulosa and 

R. monadelpha may also hybridise naturally. These hybrids resembled artificial hybrids 

obtained from crossing these varieties (DE VOS, 1972). 

Polyploidy have been induced in a number of different monocotyledons (COHEN & 

YAO, 1996; GANGA & CHEZHIYAN, 2002). As polyploidy techniques are more affective 

with propagative material obtained from micropropagation, this study will also aid in the 

establishment of a protocol to generate polyploids of Romulea species. 

The results produced in this study will be useful to consider before the design of 

experiments for the future horticultural development of species in this genus. It not only 

shows that species of this genus can be propagated, but also that there is potential for 

hybridisation and mass propagation within the genus of Romulea. 

154

Literature cited 

AHLOOWALIA, B. S. (1997). Improvement of horticultural plants through in vitro culture 

and induced mutations. Acta Horticulturae 447: 545-549. 

AHLOOWALIA, B. S., and PRAKASH, J. (2002). Physical components of tissue culture 

technology. Proceedings of a Technical Meeting organized by the Joint 

FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, Vienna.17-28. 

AHLOOWALIA, B. S., PRAKASH, J., SAVANGIKAR, V. A., and SAVANGIKAR, C. 

(2002). Plant tissue culture. Proceedings of a Technical Meeting organized by 

the Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, 

Vienna.3-10. 

ASCOUGH, G. D. (2008). Micropropagation and in vitro manipulation of Watsonia. PhD 

thesis. University of KwaZulu-Natal. 

ASCOUGH, G. D., ERWIN, J. E., and VAN STADEN, J. (2007). In vitro propagation of 

four Watsonia species. Plant Cell, Tissue and Organ Culture 88: 135–145. 

ASCOUGH, G. D., ERWIN, J. E., and VAN STADEN, J. (2008). Reduced temperature, 

elevated sucrose, continuous light and gibberellic acid promote corm formation in 

Watsonia vanderspuyiae. Plant Cell, Tissue and Organ Culture 95: 275–283. 

ASCOUGH, G. D., ERWIN, J. E., and VAN STADEN, J. (2009). Micropropagation of 

Iridaceae - a review. Plant Cell, Tissue and Organ Culture 97: 1-19. 

ASCOUGH, G. D., and FENNELL, C. W. (2004). The regulation of plant growth and 

development in liquid culture. South African Journal of Botany 70: 181-190. 




Botany 77: 216–221. 

BAIRU, M. W., STIRK, W. A., DOLEZAL, K., and VAN STADEN, J. (2007). Optimizing 

the micropropagation protocol for the endangered Aloe polyphylla: can metatopolin 

and its derivatives serve as replacement for benzyladenine and zeatin? 

Plant Cell, Tissue and Organ Culture 90: 15-23. 

BAND, S. R., and HENDRY, G. A. F. (1993). General procedures and methodologies 

In: Methods in comparative plant ecology: A laboratory manual. Eds: Grime, J.P. 

and Hendry, G.A.F. Chapman & Hall, London, pp. 252. 

155

BASKIN, C. C., and BASKIN, J. M. (1998). Seeds: Ecology, Biogeography, and 

Evolution of Dormancy and Germination. Academic Press, San Diego. 


BAXTER, B. J. M., VAN STADEN, J., GRANGER, J. E., and BROWN, N. A. C. (1994). 

Plant-derived smoke and smoke extracts stimulate seed germination of the fireclimax 

grass Themeda trianda. Environmental and Experimental Botany 34: 217- 

223. 

BEWLEY, J. D., and BLACK, M. (1982). Physiology and biochemistry of seeds in 

relation to germination. 2. 2 vols. Springer-Verlag, Berlin. 

BEWLEY, J. D., and BLACK, M. (1994). Seeds: Physiology of Development and 

Germination. Plenum Press, New York. 

BEYL, C. A. (2005). Getting started with tissue culture: Media preparation, sterile 

technique and laboratory equipment In: Plant development and biotechnology. 

Eds: Trigiano, R.N. and Gray, D.J. CRC Press LLC, Florida, pp. 19-37. 

BHAGYALAKSHMI, N. (2000). Factors influencing direct shoot regeneration from ovary 

explants of saffron. Plant Cell, Tissue and Organ Culture 58: 205–211. 

BLANK, R. R., and YOUNG, J. A. (1998). Heated substrate and smoke: Influence on 

seed emergence and plant growth. Journal of Range Management 51: 577-583. 

BRAGG, C. J., DONALDSON, J. D., and RYAN, P. G. (2005). Density of Cape 

porcupines in a semi-arid environment and their impact on soil turnover and 

related ecosystem processes. Journal of Arid Environments 61: 261–275. 

BROWN, J. T. (1990). The initiation and maintenance of callus cultures In: Plant cell 

and tissue culture. Eds: Pollard, J.W.VI VI vols Humana Press, Clifton, pp. 57-63. 

BROWN, J. T., and CHARLWOOD, B. V. (1990). Organogenesis in callus culture In: 

Plant cell and tissue culture. Eds: Pollard, J.W.VI VI vols Humana Press, Clifton, 

pp. 65-70. 

BROWN, N. A. C., and VAN STADEN, J. (1997). Smoke as a germination cue: a 

review. Plant Growth Regulation 22: 115-124. 

BURROWS, J. E., and WILLIS, C. K. (2005). Plants of the Nyika Plateau: an account of 

the vegetation of the Nyika National Parks of Malawi and Zambia. 31. Southern 

African Botanical Diversity Network, Pretoria. 

COETZEE, C., JEFTHAS, E., and REINTEN, E. (1999). Indigenous Plant Genetic 

Resources of South Africa In: Perspectives on new crops and new uses. Eds: 

Janick, J. ASHS Press, Alexandria, pp. 160-163. 

COHEN, D., and YAO, J.-L. (1996). In vitro chromosome doubling of nine Zantedeschia 

cultivars. Plant Cell, Tissue and Organ Culture 47: 43-49. 

156

COPELAND, L. O. (1976). Principles of Seed Science and Technology. Burgess 

Publishing Company, Minneapolis. 


CROSSMAN, N. D., BRYAN, B. A., and COOKE, D. A. (2008). An invasive plant and 

climate change threat index for weed risk management: Integrating habitat 

distribution pattern and dispersal process. Ecological Indicators (In press) 

DANTU, P. K., and BHOJWANI, S. S. (1987). In vitro propagation and corm formation 

in Gladiolus. Gartenbauwissenschaft 52: 90-93. 

DANTU, P. K., and BHOJWANI, S. S. (1995). In vitro corm formation and field 

evaluation of corm-derived plants of Gladiolus. Scientia Horticulturae 61: 115- 

129. 

DE BRUYN, M. H., and FERREIRA, D. I. (1992). In vitro corm production of Gladiolus 

dalennii and G.tritis. Plant Cell, Tissue and Organ Culture 31: 123-128. 

DE LANGE, J. H., and BOUCHER, C. (1990). Autecological studies on Autouinia 

capitata (Bruniaceae). Ι. Plant-derived smoke as a seed germination cue. South 

African Journal of Botany 56: 700-703. 

DE VOS, M. P. (1970a). Bydrae tot die morfologie en anatomie van Romulea: 1. Die 

knol. Journal of South African Botany 36: 215-228. 

DE VOS, M. P. (1970b). Die herontdekte Romulea monadelpha. Journal of South 

African Botany 36: 1-7. 

DE VOS, M. P. (1972). The genus Romulea in South Africa. Journal of South African 

Botany Supplement 9: 1-307. 

DE VOS, M. P. (1983). Romulea In: Flora of southern Africa. Eds: Leistner, O.A.7(2), 

fascicle 2 Botanical Research Institute, Pretoria, pp. 10-73. 

DEBERGH, P. C. A. (1994). The in vitro techniques: Their contribution to breeding and 

multiplication of ornamentals. Acta Horticulturae 353: 122-133. 

DENO, N. C. (1993). Seed Germination Theory and Practice. 2nd ed. Pensylvania State 

University Press, State College. 

DESMET, P. G. (2007). Namaqualand—A brief overview of the physical and floristic 

environment. Journal of Arid Environments 70: 570–587. 

DIXON, K. W., ROCHE, S., and PATE, J. S. (1995). The promotive effect of smoke 

derived from burnt native vegetation on seed germination of Western Australian 

plants. Oecologia 101: 185-192. 

DONAHUE, R. L., MILLER, R. W., and SHICKLUNA, J. C. (1983). Soils: An 

Introduction to Soils and Plant Growth. 5th ed. Prentice-Hall, Inc., Englewood 

Cliffs. 

157


DOWLING, P. M., CLEMENTS, R. J., and MCWILLIAM, J. R. (1971). Establishment 

and survival of a pasture species from seeds sown on the soil surface. Australian 

Journal of Agriculture Research 22: 61-74. 

EDDY, J. L., and SMITH, D. F. (1975). Seed dispersal and germination in Romulea 

rosea. Australian Journal of Experimental Agriculture and Animal Husbandry 15: 

508-512. 

EDMOND, J. B., SENN, T. L., and ANDREWS, F. S. (1964). Fundamentals of 

Horticulture. 3rd ed. McGraw-Hill, Inc., New York. 

EGERTON-WARBURTON, L. M. (1998). A smoke-induced alteration of the sub-testa 

cuticle in seeds of the post-fire recruiter, Emmenanthe penduliflora Benth. 

(Hydrophyllaceae). Journal of Experimental Botany 49: 1317–1327. 

ELLIS, R. H., and ROBERTS, E. H. (1981). The quantification of ageing and survival in 

orthodox seeds. Seed Science and Technology 9: 373-409. 

EMERY, N., and ATKINS, C. (2006). Cytokinins and seed development In: Handbook 

of Seed Science and Technology. Eds: Basra, A.S. The Haworth Press, Inc., New 

York, pp. 63-93. 

EVANS, R. A., ECKERT, E. R., and KAY, B. L. (1967). Wheatgrass establishment with 

paraquat and tillage on Downy Brome range. Weeds 15: 50–55. 

FINER, J. J. (1995). Direct somatic embryogenesis In: Plant cell, tissue and organ 

culture: Fundamental methods. Eds: Gamborg, O.L. and Phillips, G.C. Springer- 

Verlag, New York, pp. 91-101. 

FLAISHMAN, M. A., and KAMENETSKY, R. (2006). Florogenesis in flower bulbs: 

classical and molecular approaches In: Floriculture, Ornamental and Plant 

Biotechnology: Advances and Topical Issues. Eds: Teixera Da Silva, J.A. Global 

Science Books, UK, pp. 33-43. 

FLEMATTI, G. R., GHISALBERT, E. L., DIXON, K. W., and TRENGOVE, R. D. (2004). 

Molecular weight of a germination-enhancing compound in smoke. Plant and Soil 

263: 1–4. 

FLEMING, K. M., LOPER, R., and PRIVETT, V. (2009). Romulea rosea (Iridaceae): 

adventive in Texas. Phytologia 91: 73-75. 

GABA, V. P. (2004). Plant growth regulators in plant tissue culture and development In: 

Plant development and biotechnology. Eds: Trigiano, R.N. and Gray, D.J. CRC 

Press, New York, pp. 97-99. 

GAMBORG, O. L., and PHILLIPS, G. C. (1995). Sterile techniques In: Plant cell, tissue 

and organ culture: Fundamental methods. Eds: Gamborg, O.L. and Phillips, G.C. 

Springer-Verlag, New York, pp. 35-41. 

158


GANGA, M., and CHEZHIYAN, N. (2002). Influence of the antimitotic agents colchicine 

and oryzalin on in vitro regeneration and chromosome doubling of diploid 

bananas (Musa spp.). Journal of Horticultural Science & Biotechnology 77: 572- 

575. 

GARDNER, C. M. K., BELL, J. P., COOPER, J. D., DEAN, T. J., HODNETT, M. G., 

and GARDNER, N. (1991). Soil water content In: Soil Analysis: Physical 

Methods. Eds: Smith, K.A. and Mullins, C.E. Marcel Dekker, Inc., New York, pp. 

620. 

GAUCH, H. G. (1972). Inorganic Plant Nutrition. Dowden, Hutchinson & Ross, Inc., 

Stroudsburg. 

GEORGE, E. F., and SHERRINGTON, P. D. (1984). Plant Propagation by Tissue 

Culture: Handbook and directory of commercial laboratories. Exegetics, 

Basingstoke. 

GINZBURG, C., and ZIV, M. (1973). Hormonal regulation of cormel formation in 

Gladiolus stolons grown in vitro. Annals of Botany 37: 219-224. 

GLASS, A. D. M. (1989). Plant Nutrition: An Introduction to Current Concepts. Jones 

and Bartlett Publishers, Inc., Boston. 

GOLDBLATT, P. (1990). Phylogeny and classification of Iridaceae. Annals of the 

Missouri Botanical Garden 77: 607-627. 

GOLDBLATT, P. (1991). African Iridaceae In: Systematics, Biology and Evolution of 

some South African taxa. Eds: Linder, H.P. and Hall, A.V. Bolus Herbarium, Cape 

Town, pp. 1-74. 

GOYAL, K., WALTON, L. J., and TUNNACLIFFE, A. (2005). LEA proteins prevent 

protein aggregation due to water stress. Biochemical Journal 388: 151–157. 

GRAY, D. J. (2005). Propagation from nonmeristematic tissues: Nonzygotic 

embryogenesis In: Plant development and biotechnology. Eds: Trigiano, R.N. 

and Gray, D.J. CRC Press, New York, pp. 187-200. 

HARPER, J. L., and BENTON, R. A. (1966). The behaviour of seeds in soil. II. The 

germination of seeds on the surface of a water supplying substrate. Journal of 

Ecology 54: 151–166. 

HARTMANN, H. T., and KESTER, D. E. (1965). Plant propagation: Principles and 

Practices. 2nd ed. Prentice-Hall, Inc., New Jersey. 

HEWITT, E. J., and SMITH, T. A. (1974). Plant Mineral Nutrition. The English 

Universities Press Ltd., London. 

159


HILTON-TAYLOR, C. (1996). Red Data List of Southern African Plants. Strelitzia, 4. 

National Botanical Institute, Pretoria. 

HOMES, J., LEGROS, M., and JAZIRI, M. (1987). In vitro multiplication of Crocus 

sativus L. Acta Horticulturae 212: 675-676. 

HU, C. Y., and ZANETTINI, M. H. B. (1995). Embryo culture and embryo rescue for 

wide cross hybrids In: Plant cell, tissue and organ culture: Fundamental 

methods. Eds: Gamborg, O.L. and Phillips, G.C. Springer-Verlag, New York, pp. 

129-140. 

HUSSEY, G. (1975). Totipotency in tissue explants and callus of some members of the 

Liliaceae, Iridaceae, and Amaryllidaceae. Journal of Experimental Botany 26: 

253-262. 

HUSSEY, G. (1976). In vitro release of axilary shoots from apical dominance in 

monocotyledonous plantlets. Annals of Botany 40: 1323-1325. 

HUSSEY, G. (1977a). In vitro propagation of Gladiolus by precocious axillary shoot 

formation. Scientia Horticulturae 6: 287-296. 

HUSSEY, G. (1977b). In vitro propagation of some members of the Liliaceae, Iridaceae 

and Amaryllidaceae. Acta Horticulturae 78: 303-309. 

HUYLENBROECK, J. M., PIQUERAS, A., and DEBERGH, P. C. (1998). 

Photosynthesis and carbon metabolism in leaves formed prior and during ex vitro 

acclimatization of micropropagated plants. Plant Science 134: 21–30. 

HUYLENBROECK, J. M., PIQUERAS, A., and DEBERGH, P. C. (2000). The evolution 

of photosynthetic capacity and the antioxidantenzymatic system during 

acclimatization of micropropagated Calathea plants. Plant Science 155: 59–66. 

IMANISHI, H. (1983). Effects of exposure of bulbs to smoke and ethylene on flowering 

of Narcissus tazetta cultivar ‘Grand Soleil d'Or’. Scientia Horticulturae 21: 173- 

180. 

INTERNATIONAL SEED TESTING ASSOCIATION (1999a). Biochemical test for 

viability. Seed Science and Technology 27, Supplement: 33-35. 

INTERNATIONAL SEED TESTING ASSOCIATION (1999b). Excised embryo test for 

viability. Seed Science and Technology 27, Supplement: 65-67. 

JÄGER, A. K., LIGHT, M. E., and VAN STADEN, J. (1996). Effects of source of plant 

material and temperature on the production of smoke extracts that promote 

germination of light-sensitive lettuce seeds. Environmental and Experimental 

Botany 36: 421-429. 

JÄGER, A. K., MCALISTER, B. G., and VAN STADEN, J. (1995). In vitro culture of 

Babiana species. Acta Horticulturae 420: 107-108. 

160


JÄGER, A. K., MCALISTER, B. G., and VAN STADEN, J. (1998). In vitro culture of 

Gladiolus carneus. South African Journal of Botany 64: 146-147. 

JANN, R. C., and AMEN, R. D. (1977). What is germination? In: The physiology and 

biochemistry of seed dormancy and germination. Eds: Khan, A.A. North-Holland, 

New York, pp. 7-28. 

JONES, R. L., and STODDART, J. L. (1977). Gibberellins and seed germination In: 

The Physiology and Biochemistry of Seed Dormancy and Germination. Eds: 

Khan, A.A. North-Holland, New York, pp. 78-109. 

KANE, M. E. (2004). Shoot culture procedures In: Plant development and 

biotechnology. Eds: Trigiano, R.N. and Gray, D.J. CRC, New York, pp. 145-157. 

KHAN, A. A. (1977). Seed dormancy: changing concepts and theories In: The 

Physiology and Biochemistry of Seed Dormancy and Germination. Eds: Khan, 

A.A. North-Holland, New York, pp. 29-50. 

KOETLE, M. J., FINNIE, J. F., and VAN STADEN, J. (2010). In vitro regeneration in 

Dierama erectum. Plant Cell, Tissue and Organ Culture 103: 23-31. 

KÖK, T., BILGEN, A., ÖZDEMIR, C., KUTBAY, H. G., and KESKIN, M. (2007). 

Macroelement (N, P, K) contents of Romulea columnae Seb. and Mauri Subsp. 

culumnae during vegetative and generative growth phases. Journal of Plant 

Sciences 2: 440-446. 

KOLLER, D. (1972). Environmental Control of Seed Germination In: Seed Biology. Eds: 

Kozlowski, T.T.2 3 vols Academic Press, Inc., New York, pp. 447. 

KRIKORIAN, A. D., and KANN, R. P. (1986). Regeneration in Liliaceae, Iridaceae, and 

Amaryllidaceae In: Cell Culture and Somatic Cell Genetics of Plants. Eds: Vasil, 

I.K.3 Academic Press, Orlando, pp. 187-205. 

KULKARNI, M. G., SPARG, S. G., and VAN STADEN, J. (2006a). Dark conditioning, 

cold stratification and a smoke-derived compound enhance the germination of 

Eucomis autumnalis subsp. autumnalis seeds. South African Journal of Botany 

72: 157–162. 

KULKARNI, M. G., SPARG, S. G., and VAN STADEN, J. (2006b). Germination and 

post-germination response of Acacia seeds to smoke-water and butenolide, a 

smoke-derived compound. Journal of Arid Environments 69: 177–187. 

KULKARNI, M. G., STREET, R. A., and VAN STADEN, J. (2007). Germination and 

seedling growth requirements for propagation of Dioscorea dregeana (Kunth) 

Dur. and Schinz — A tuberous medicinal plant. South African Journal of Botany 

73: 131–137. 

161


KUMAR, A., SOOD, A., PALNI, L. M. S., and GUPTA, A. K. (1999). In vitro propagation 

of Gladiolus hybridus Hort.: Synergistic effect of heat shock and sucrose on 

morphogenesis. Plant Cell, Tissue and Organ Culture 57: 105-112. 

KYTE, L., and KLEYN, J. (1996). Plants from test tubes: An introduction to 

micropropagation. 3rd ed. Timber Press, Inc., Portland. 

LAWRENCE, E. (2000). Henderson's dictionary of biological terms. 12 ed. Prentice Hall, 

Harlow. 

LEADEM, C. L. (1984). Quick Tests for Tree Seed Viability. 18. B.C. Ministry of Forests 

Research Laboratory, Victoria. 

LECLERC, J.-C. (2003). Plant Ecophysiology. Science Publishers, Inc., Enfield. 

LEUBNER-METZGER, G. (2006). Hormonal interactions during seed dormancy release 

and germination In: Handbook of Seed Science and Technology. Eds: Basra, 

A.S. The Haworth Press, Inc., New York, pp. 303-341. 

LIGHT, M. E., GARDNER, M. J., JÄGER, A. K., and VAN STADEN, J. (2002). Dual 

regulation of seed germination by smoke solutions. Plant Growth Regulation 37: 

135-141. 

LIGHT, M. E., KULKARNI, M. G., ASCOUGH, G.D., and VAN STADEN, J. (2007). 

Improved flowering of a South African Watsonia with smoke treatments. South 

African Journal of Botany 73: 298 

LILIEN-KIPNIS, H., and KOCHBA, M. (1987). Mass propagation of new gladiolus 

hybrids. Acta Horticulturae 212: 631-637. 

LOVELAND, P. J., and WHALLEY, W. R. (1991). Particle size analysis In: Soil 

Analysis: Physical Methods. Eds: Smith, K.A. and Mullins, C.E. Marcel Dekker, 

Inc., New York, pp. 620. 

MADUBANYA, L. G. (2004). In vitro conservation of endangered Dierama species. MSc 

thesis. University of KwaZulu-Natal. 

MANNING, J. C., and GOLDBLATT, P. (1996). West coast. South African Wild Flower 

Guide, 7. Botanical Society of South Africa, Kirstenbosch. 

MANNING, J. C., and GOLDBLATT, P. (1997). Nieuwoudtville: Bokkeveld Plateau & 

Hantam. South African Wild Flower Guide, 9. Botanical Society of South Africa, 

Kirstenbosch. 

MANNING, J. C., and GOLDBLATT, P. (2001). A synoptic review of Romulea 

(Iridaceae: Crocoideae) in sub-Saharan Africa, the Arabian Peninsula and 

Socotra including new species, biological notes, and a new infrageneric 

classification. Adansonia 23: 59-108. 

162


MARION-POLL, A. (1997). ABA and seed development. TRENDS in Plant Science 2: 

447-448. 

MARSCHNER, H. (1995). Mineral Nutrition of Higher Plants. Second ed. Academic 

Press Ltd., London. 

MAYER, A. M. (1977). Metabolic control of germination In: The physiology and 

biochemistry of seed dormancy and germination. Eds: Khan, A.A. North-Holland, 

New York, pp. 357-384. 

MCALISTER, B. G., JÄGER, A. K., and VAN STADEN, J. (1998). Micropropagation of 

Babiana spp. South African Journal of Botany 64: 88-90. 

MEEKLAH, F. A. (1958). Chemical aids to pasture renovation. Proceedings of 20th 

Conference of the New Zealand Grassland Association, pp. 116–119. 

MEYER, H. J., and VAN STADEN, J. (1988). In vitro multiplication of Ixia flexuosa. 

Horticultural Science 23: 1070-1071. 

MILLER, H. P., and PERRY, R. A. (1968). Preliminary studies on the establishment of 

Townsville Lucerne (Stylosanthes humilis) in uncleared native pasture at 

Katherine N.T. Australian Journal of Experimental Agriculture and Animal 

Husbandry 8: 26–32. 

MINEO, L. (1990). Plant tissue culture techniques In: Tested studies for laboratory 

teaching. Eds: Goldman, C.A.11 ABLE, Pennsylvania, pp. 151-174. 

MINORSKY, P. V. (2002). The Hot and the Classic. Plant Physiology 128: 1167–1168. 

MURASHIGE, T., and SKOOG, F. (1962). A revised medium for rapid growth and 

bioassay with tobacco tissue cultures. Physiologia Plantarum 15: 472-497. 

NIEDERWIESER, J. G., KLEYNHANS, R., and HANCKE, F. L. (2002). Development of 

a new flower bulb crop in South Africa. Acta Horticulturae 570: 67-71. 

NOWAK, J., and PRUSKI, K. (2002). Priming tissue cultured propagules. Proceedings 

of a Technical Meeting organized by the Joint FAO/IAEA Division of Nuclear 

Techniques in Food and Agriculture, Vienna.69-82. 

PAGE, Y. M., and VAN STADEN, J. (1985). In vitro propagation of Dierama latifolium. 

Horticultural Science 20: 1049-1050. 

PENG, J., and HARBERD, N. P. (2002). The role of GA-mediated signalling in the 

control of seed germination. Current Opinion in Plant Biology 5: 376–381. 

PHILLIPS, G. C., HUSTENBERGER, J. F., and HANSEN, E. E. (1995). Plant 

regeneration by organogenesis from callus and cell suspension cultures In: Plant 

cell, tissue and organ culture: Fundamental methods. Eds: Gamborg, O.L. and 

Phillips, G.C. Springer-Verlag, New York, pp. 35-42. 

163


PIERIK, R. L. M. (1997). In Vitro Culture of Higher Plants. Kluwer Academic Publishers, 

Dordrecht. 

PLESSNER, O., ZIV, M., and NEGBI, M. (1990). In vitro corm production in the saffron 

crocus (Crocus sativus L.). Plant Cell, Tissue and Organ Culture 20: 89-94. 

PONTING, M. (2006). Wild flowers of the Cape and Namaqualand. Tour Report. 

Naturetrek, England. 

RAGHAVAN (1976). Experimental embryogenesis in vascular plants. Experimental 

Botany (Sutcliffe, J.F.).(eds), 10. 10 vols. Academic Press, London. 

RAIMONDO, D., VON STADEN, L., FODEN, W., VICTOR, J. E., HELME, N. A., 

TURNER, R. C., KAMUNDI, D. A., and MANYAMA, P. A. (2009). Red List of 

South African Plants 2009. Strelitzia 25. South African National Biodiversity 

Institute, Pretoria. 

REED, S. M. (2005). Embryo rescue In: Plant development and biotechnology. Eds: 

Trigiano, R.N. and Gray, D.J. CRC Press, New York, pp. 235-239. 

REEVES, G., CHASE, M. W., GOLDBLATT, P., RUDALL, P., FAY, M. F., COX, A. V., 

LEJEUNE, B., and SOUZA-CHIES, T. (2001). Molecular systematics of 

Iridaceae: Evidence from four plastid DNA Regions. American Journal of Botany 

11: 2074–2087. 

REINTEN, E., and COETZEE, J. H. (2002). Commercialization of South African 

indigenous crops: Aspects of research and cultivation of products In: Trends in 

New Crops and New Uses. Eds: Janick, J. and Whipkey, A. ASHS Press, 

Alexandria, VA, pp. 76-80. 

REMOTTI, P. C., and LÖFFLER, H. J. M. (1995). Callus induction and plant 

regeneration from gladiolus. Plant Cell, Tissue and Organ Culture 42: 171-178. 

ROBERTS, A. V., SMITH, E. F., and MOTTLEY, J. (1990). The preparation of 

micropropagated plantlets for transfer to soil without acclimatization In: Plant cell 

and tissue culture. Eds: Pollard, J.W.VI VI vols Humana Press, Clifton, pp. 227- 

236. 

SARAH, P. (2004). Soil sodium and potassium adsorption ratio along a Mediterranean– 

arid transect. Journal of Arid Environments 59: 731–741. 

SARASAN, V., CRIPPS, R., RAMSAY, M. M., ATHERTON, C., MCMICHEN, M., 

PRENDERGAST, G., and ROWNTREE, J. K. (2006). Conservation in vitro of 

threatened plants: Progress in the past decade In Vitro Cellular & Developmental 

Biology 42: 206-214. 

SCHWARTZ, O. J., SHARMA, R. A., and BEATY, R. M. (2004). Propagation from 

nonmeristematic tissues: Organogenesis In: Plant development and 

164


biotechnology. Eds: Trigiano, R.N. and Gray, D.J. New York, CRC Press, pp. 

159-172. 

SEDGLEY, R. H. (1963). The importance of liquid-seed contact during the germination 

of Medicago tribuloides Desr. Australian Journal of Agriculture Research 14: 

646–653. 

SEN, J., and SEN, S. (1995). Two-step bud culture technique for high frequency 

regeneraion of Gladiolus corms. Scientia Horticulturae 64: 133-138. 

SENGUPTA, J., and SEN, S. (1988). Karyological analysis of cultured cells and 

regenerated plants of Cipura paludosa Aubl. (Iridaceae) - A structural hybrid. 

Bulletin of the Torrey Botanical Club 115: 280-289. 

SHIBLI, R. A., SHATNAWI, M. A., SUBAIH, W. S., and AJLOUNI, M. M. (2006). In 

Vitro Conservation and Cryopreservation of Plant Genetic Resources: A Review. 

World Journal of Agricultural Sciences 2: 372-382. 

SIMMONDS, H., HOLST, P., and BOURKE, C. (2000). The palatability, and potential 

toxicity of Australian weeds to goats. Rural Industries Research and Development 

Corporation, Kingston. 

SLABBERT, M. M., DE BRUYN, M. H., FERREIRA, D. I., and PRETORIUS, J. (1993). 

Regeneration of bulblets from twin scales of Crinum macowanii in vitro. Plant 

Cell, Tissue and Organ Culture 33: 133-141. 

SLATER, A., SCOTT, N. W., and FOWLER, M. R. (2003). Plant Biotechnology: The 

Genetic Manipulation of Plants. Oxford University Press, New York. 

SLEEMAN, J. R., and BREWER, R. (1988). Soil Structure and Fabric. CSIRO, 

Adelaide. 

SMITH, J. (2000a). Micro-propagation of the Gymea Lily. Rural Industries Research and 

Development Corporation, Sydney. 

SMITH, R. H. (2000b). Plant Tissue Culture: Techniques and Experiments. 2 nd edition 

ed. Academic Press, London. 

SOIL CLASSIFICATION WORKING GROUP (1991). Soil Classification: A Taxonomic 

System for South Africa. Department of Agricultural Development, Pretoria. 

SPARG, S. G., KULKARNI, M. G., LIGHT, M. E., and VAN STADEN, J. (2005). 

Improving seedling vigour of indigenous medicinal plants with smoke. 

Bioresource technology 96: 1323-1330. 

SPARG, S. G., KULKARNI, M. G., and VAN STADEN, J. (2006). Aerosol smoke and 

smoke-water stimulation of seedling vigor of a commercial maize cultivar. Crop 

Science 46: 1336-1340. 

165


STEINITZ, B., COHEN, A., GOLDBERG, Z., and KOCHBA, M. (1991). Precocious 

gladiolus corm formation in liquid shake cultures. Plant Cell, Tissue and Organ 

Culture 26: 63-70. 

STEINITZ, B., and LILIEN-KIPNIS, H. (1989). Control of precocious Gladiolus corm and 

cormel formation in tissue culture. Journal of Plant Physiology 135: 495-500. 

STEPHAN-SARKISSIAN, G. (1990). Selection of media for tissue and cell culture In: 

Plant cell and tissue culture. Eds: Pollard, J.W.VI VI vols Humana Press, Clifton, 

pp. 13-26. 

SUTTER, E. G. (1986). Micropropagation of Ixia viridifolia and a Gladiolus x 

Homoglossum hybrid. Scientia Horticulturae 29: 181-189. 

TAN NHUT, D., TEIXEIRA DA SILVA, J. A., XUAN HUYEN, P., and PAEK, K. Y. 

(2004). The importance of explant source on regeneration and micropropagation 

of Gladiolus by liquid shake culture. Scientia Horticulturae 102: 407-414. 

TAYLOR, J. L. S., and VAN STADEN, J. (1998). Plant-derived smoke solutions 

stimulate the growth of Lycopersicon esculentum roots in vitro. Plant Growth 

Regulation 26: 77-83. 

THIMANN, K. V. (1977). Prefactory chapter In: The Physiology and Biochemistry of 

Seed Dormancy and Germination. Eds: Khan, A.A. North-Holland, New York, pp. 

1-28. 

TIAN, X., KNAPP, A. D., GIBSON, L. R., STRUTHERS, R., MOORE, K. J., 

BRUMMER, E. C., and BAILEY, T. B. (2003). Response of Eastern Gamagrass 

Seed to Gibberellic Acid Buffered below Its pKa. Crop Science 43: 927–933. 

TILLICH, H. J. (2003). Seedling morphology in Iridaceae: Indications for relationships 

within the family and to related families. Flora: 220–242. 

UYEMURA, S., and IMANISHI, H. (1984). Effects of duration of exposure to ethylene on 

dormancy release in freesia corms. Scientia Horticulturae 22: 383-390. 

VAN KLEUNEN, M., MANNING, J. C., PASQUALETTO, V., and JOHNSON, S. D. 

(2008). Phylogenetically independent associations between autonomous selffertilization 

and plant invasiveness. The American Naturalist 171: 195-201. 

VAN STADEN, J., JÄGER, A. K., LIGHT, M. E., and BURGER, B. V. (2004). Isolation 

of the major germination cue from plant-derived smoke. South African Journal of 

Botany 70: 654–659. 

WEST, A. G. (2009). Climate change, drought and biodiversity: an ecophysiological 

perspective. Thirty-fifth Annual conference of the South African Association of 

Botanists (SAAB), 19 to 22 January, University of Johannesburg, Drakensberg. 

166

WITHERS, L. A. (2008). In vitro conservation. Biological Journal of the Linnaean 

Society 43: 31-42. 

WOCHOCK, Z. S. (1981). The role of tissue culture in preserving threatened and 

endangered plant species. Biological Conservation 20: 83-89. 


YABUYA, T., YOSHIHARA, N., INOUE, K., and SHIMIZU, H. (2006). Advances in 

breeding of Japanese garden iris. Floriculture, Ornamental and Plant 

Biotechnology 1: 556-563. 

YASMIN, S., AHMED, B., and SOOMRO, R. (2003). Influence of ABA, Gibberellin and 

Kinetin on IAA induced adventitious root development on hypocotyl cuttings of 

mungbean. Biotechnology 2: 37-43. 

ZIV, M. (1989). Enhanced shoot and cormlet proliferation in liquid cultured gladiolus 

buds by growth retardants. Plant Cell, Tissue and Organ Culture 17: 101-110. 

ZIV, M. (1997). The contribution of biotechnology to breeding, propagation and disease 

resistance in geophytes. Acta Horticulturae 430: 247-258. 

ZIV, M., HALEVY, A. H., and SHILO, R. (1970). Organs and plantlets regeneration of 

Gladiolus through tissue culture. Annals of Botany 34: 671-676. 

ZIV, M., and LILIEN-KIPNIS, H. (2000). Bud regeneration from inflorescence explants 

for rapid propagation of geophytes in vitro. Plant Cell Reports 19: 845-850. 

ZIV, M., RONEN, G., and RAVIV, M. (1998). Proliferation of meristematic clusters in 

disposable presterilised plastic bioreactors for the large-scale micropropagation of 

plants. In Vitro Cellular & Developmental Biology 34: 152-158. 

167

View/Open - ResearchSpace - University of KwaZulu-Natal

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?