The reproductive system of Campuloclinium macrocephalum and its 

implications for biocontrol implementation. 

by 

Saness Moodley 

(1466053) 

 

Dissertation 

Submitted in fulfilment of the requirements for the degree 

Master of Science 

in 

Animal, Plant and Environmental Sciences 

in the Faculty of Science, University of the Witwatersrand, Johannesburg, South Africa 

 

Supervisor: Professor Glynis Goodman-Cron 

Co-supervisor: Dr Kelsey Glennon 

July 2022 



ii 
 

Declaration  
 

I declare that this dissertation is my own, unaided work. It has been submitted in fulfilment of 

the requirements of a Master of Science at the University of the Witwatersrand. It has not been 

submitted before for any degree or examination to another university or similar institution.   

 

 

Saness Moodley 

8 July 2022 

 

 

 

 

 

 

Supervisors: 

Professor Glynis Goodman-Cron 

Dr Kelsey Glennon 

 

 

 

 

 



iii 
 

Table of contents   
 

DECLARATION ............................................................................................................................................ II 

TABLE OF CONTENTS ..............................................................................................................................III 

ABSTRACT ................................................................................................................................................... V 

ACKNOWLEDGMENTS .......................................................................................................................... VIII 

INTRODUCTORY CHAPTER ..................................................................................................................... 1 

RATIONALE ........................................................................................................................................................ 1 

SEXUAL REPRODUCTIVE STRATEGIES ..................................................................................................... 2 

ASEXUAL REPRODUCTIVE STRATEGIES .................................................................................................. 5 

THE CO-OCCURRENCE OF APOMIXIS AND POLYPLOIDY ......................................................................... 8 

MALE FERTILITY, POLYPLOIDY AND APOMIXIS ................................................................................... 10 

CAMPULOCLINIUM MACROCEPHALUM ................................................................................................. 11 

AIM AND OBJECTIVES ...................................................................................................................................... 14 

DISSERTATION OUTLINE .................................................................................................................................. 15 

REFERENCES ............................................................................................................................................... 18 

CHAPTER 2 ................................................................................................................................................. 32 

ABSTRACT .................................................................................................................................................... 32 

INTRODUCTION .......................................................................................................................................... 33 

METHOD AND MATERIALS ..................................................................................................................... 37 

RESULTS ....................................................................................................................................................... 42 

DISCUSSION ................................................................................................................................................. 50 

CONCLUSION ............................................................................................................................................... 53 

REFERENCES ............................................................................................................................................... 54 

CHAPTER 3 ................................................................................................................................................. 65 

ABSTRACT .................................................................................................................................................... 65 

INTRODUCTION .......................................................................................................................................... 66 

METHOD AND MATERIALS ..................................................................................................................... 70 

RESULTS ....................................................................................................................................................... 78 

DISCUSSION ................................................................................................................................................. 95 

CONCLUSION ............................................................................................................................................. 103 

REFERENCES ............................................................................................................................................. 104 

CONCLUDING CHAPTER ...................................................................................................................... 119 



iv 
 

GENERAL OVERVIEW .................................................................................................................................... 119 

SUMMARY OF THE STUDY AND MAJOR FINDINGS.......................................................................................... 121 

FUTURE RECOMMENDATIONS ....................................................................................................................... 125 

CONCLUDING REMARKS ................................................................................................................................ 127 

REFERENCES ............................................................................................................................................. 127 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

 

 

 

 

 

 

 

 



v 
 

Abstract 

 

Invasive species are a threat to biodiversity therefore it is imperative to determine the factors 

that facilitate the invasion potential of a species. Campuloclinium macrocephalum Less. (DC), 

the ‘pompom weed’, is an alien invasive species in South Africa and is currently threatening 

the persistence of the grassland, wetland, and savanna biomes. The species is also significantly 

contributing to a decline in plant diversity by outcompeting native vegetation in these areas. 

Various integrated approaches using combinations of chemical, mechanical and biocontrol 

management programs have been developed to manage the spread of the species in its invaded 

range, however the species has still been able to persist.  

The persistence of the species was hypothesised to be a consequence of the co-occurrence of 

apomixis and polyploidy, however despite the identification of triploid and tetraploid cytotypes 

in South African populations of the pompom weed, the reproductive strategy of the species has 

not yet been determined. The aim of this study was therefore to infer whether populations of 

C. macrocephalum (pompom weed) reproduces via vector-mediated crosses, self-pollination 

or apomixis (either facultative or obligate) and examine the relationship of the mode of 

reproduction with ploidy level. Male fertility was also assessed to ascertain if interploidy gene 

flow was possible. The collated information was then used to infer the potential impact of 

reproductive strategies and polyploidy on biocontrol. All examined populations were shown to 

have high mean pollen viability percentages of 90% and 98% with no significant differences 

in pollen viability amongst the four populations. The high pollen viability percentages were 

supported by prolific pollen grain germination on the stigmatic surfaces (margins of style at 

base of style branches) and the sides of the style. This suggested that the pollen grains can 

fertilize and interploidy mating is likely possible in South African populations of the pompom 

weed. It is plausible that the high pollen viability is enabling triploids to act as a ‘triploid 



vi 
 

bridge’. However, the high pollen viability was confounded by the pollen tube analyses 

revealing that pollen tube growth is being arrested on the stigmatic surface suggesting that 

overall male fertility is low. The arrested pollen tube growth is typically associated with a 

‘triploid block’. Nevertheless, the production of viable gametes can reduce the triploid block 

and facilitate gene flow between populations.  

The predominant mode of reproduction was determined by assessing the contribution of insects 

to pollen transfer, pollinator exclusion experiments, germination trials, pollen tube growth to 

the ovules and genetic analyses. We found that the African Monarch butterfly (Danaus 

chrysippus) and the honeybee (Apis mellifera) contributed most to pollen transfer in 

comparison to the other insects visiting C. macrocephalum. A pollinator exclusion experiment 

showed that the pompom weed can set seed in the absence of pollinators, albeit at lower 

quantities than in the open treatments. Nevertheless, germination percentages showed that 

reproductive success was similar between open and bagged treatments in each population. The 

Modderfontein population showed lower reproductive success and seedling establishment in 

comparison to the other populations, presumably due to the severity of the biocontrol 

infestation on this population. Genetic analyses revealed low genetic variation within and 

amongst populations. Pollen tube analyses showed no pollen tube growth to the ovules in all 

samples, which suggests that seed set is independent of fertilization. The lack of pollen tube 

growth is a strong indicator of obligate autonomous apomixis which is further corroborated by 

the low genetic differentiation between maternal plants and their respective offspring. 

The co-occurrence of apomixis and polyploidy made it difficult to discern which factor 

contributes more to the invasiveness of the species, however, we hypothesise that autonomous 

apomixis provides the pompom weed with the competitive advantage to persist in its invaded 

range. However, further studies on the reproductive strategies of tetraploid cytotypes are 

needed to confirm this hypothesis. The low genetic variation suggests that all populations 



vii 
 

should be equally susceptible to biocontrol agents, however that this may be affected by other 

factors such as environmental conditions or phenotypic plasticity. Phenotype plasticity refers 

to a single genotype producing different phenotypes in response to environmental conditions. 

This could reduce the efficacy of biocontrol agents as they may exhibit differential responses 

on different phenotypes. 

Keywords: Invasive species; Autonomous apomixis; Polyploidy; Male fertility; Biocontrol 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



viii 
 

Acknowledgments 

 

This project was funded by the University of the Witwatersrand (Postgraduate Merit Award) 

and the National Research Foundation (Reference: MND200618533709). Thank you to both 

parties for the opportunity and for enabling me to make a meaningful contribution to science. 

I would like to thank my supervisors Professor Glynis Goodman-Cron and Dr Kelsey Glennon. 

I truly appreciate all your editorial (amongst other) feedback, assistance with lab work, support, 

and encouragement. You both continuously inspire me, and I am truly grateful to have had the 

opportunity to work with both of you. 

Thank you to Dr Kenneth Oberlander and Sinethemba Ntshangase from the University of 

Pretoria for allowing me to use the flow cytometer. A special thank you to Sinethemba, who 

ran my samples for me and assisted with data analyses. Many thanks to the Microscopy and 

Microanalysis Unit at the University of the Witwatersrand, specifically Dr Deran Reddy and 

Jacques Gerber, for their assistance with the fluorescence microscopy and scanning electron 

microscopy.  

To my mom, Edal Naidoo, thank you for always ensuring that I ‘reach for the stars’ despite all 

the challenges we have had to face. You never compromised on my education and ensured that 

I would have the opportunity to study further. Your faith in me sustains me through all of life’s 

difficulties. Thank you to my brother, Ruveshen Moodley, for always going above and beyond 

the role of a sibling. All the sacrifices you make to ensure that I can follow my dreams do not 

go unnoticed. I am truly grateful to have a sibling who believes in me and encourages me as 

much as you do. 

To the PEAS lab, thank you all for the encouraging words and support. The completion of this 

thesis would not have been possible without Promise Mtileni, Hendrik Niemann, Thando 



ix 
 

Twala, Timothy Hall, and Bianca Ferreira. Promise and Hendrik, thank you for all your 

assistance with my flow cytometry and for always encouraging me on my ‘down days’. 

Thando, thank you for supporting me from the beginning of this journey. You were never too 

busy to answer my endless questions or provide words of encouragement. Timothy Hall, thank 

you for never hesitating to help me, your kindness is greatly appreciated. Bianca Ferreira, thank 

you for all your assistance in the lab from extracting DNA to loading gels – you were always 

on call to share your knowledge. It is rare to find a person that is so kind and always willing to 

help. Thank you for the chats and coffee runs, they helped me get through all the tough lab 

days! 

A heartfelt thank you to Jessica Minnaar, who always made time to support and encourage me. 

You are a phenomenal scientist and an even better person. To my good friend, Verona 

Govender, thank you for all your assistance in the lab, for reading my drafts, and for simply 

being there. Thank you for all the laughs and smiles along the way, I am so grateful to be 

sharing my postgraduate journey with you. Thank you to Riyanta Naidoo and Yashila 

Govender for all their assistance with my data analyses. I am truly blessed to have such kind 

friends.  

To all my loved ones in heaven, I wish you were here to celebrate this milestone with me. To 

my Dean Daddy, thank you for your unwavering support through the years. You were my best 

friend and I miss you every day. To my Aum ma, thank you for always praying for me, it gave 

me the strength to persevere. To my Uncle Myen, Aunty Kim and Uncle Yogan, thank you for 

always cheering me on and for making sure that I knew how proud you were of me. Losing 

you all so close to the submission of this thesis was incredibly difficult, but I hope that I have 

made you proud. 

 



1 
 

Introductory chapter 
 

Rationale 

 

Plant invasions threaten biodiversity and ecosystem function by outcompeting native 

vegetation (Pimentel et al., 2001; Hawkes, 2007). Invasive species typically invade agricultural 

fields and natural areas, reduce productivity and biodiversity respectively, and alter interactions 

amongst native species (Holzmueller and Jose, 2009). Additionally, the sustainability of native 

communities is further threatened by the structural, functional, and compositional changes that 

follow a successful plant invasion (Holzmueller and Jose, 2009).  

Campuloclinium macrocephalum (Less) DC., colloquially referred to as the ‘pompom weed’, 

is native to South and Central America but is an invasive species in South Africa (Henderson, 

2001; Henderson, 2007). Predicted distributions suggest that as the species expands its range, 

it poses a considerable risk to the conservation of the savanna and grassland biomes (Trethowan 

et al., 2011). The species outcompetes native vegetation thereby causing a significant decline 

in plant diversity (Aileen, 2005). Despite the use of various mechanical, chemical and 

biocontrol methods, pompom weed has persisted.  

The success of C. macrocephalum was hypothesised to be linked to the co-occurrence of 

polyploidy and apomixis (Gitonga et al., 2015). Polyploid apomictics are formidable invaders 

(Richards, 2003), however the association between polyploidy and apomixis is rarely 

considered in biocontrol management plans. Both polyploidy and apomixis have implications 

for progeny formation and the subsequent genetic diversity within progeny – factors which 

may facilitate invasion success (Krahulcová and Krahulec, 2021). Triploid and tetraploid 

cytotypes have been identified in South African populations of C. macrocephalum (Gitonga et 

al., 2022), however there is no clear consensus on what reproductive strategy the species uses. 



2 
 

Research on the reproductive strategies employed by an invasive species reveals important 

information on their invasive mechanisms and enables us to develop strategies to mitigate the 

spread and effects of these species (Yan et al., 2016).  

Sexual reproductive strategies  

 

In plants, sexual reproduction is mediated by the transfer of pollen from the anthers to the 

stigma using wind, water, mammals, birds, insects, or gravity (Richards, 1996; Abrol, 2012). 

Cross-pollination is the transfer of pollen between individuals on different plants, whereas 

geitonogamy occurs when the pollen transfer is between individual flowers on the same plant 

and autogamy is pollen transfer within the same flower (Charlesworth, 2006; Abrol, 2012). 

Autogamy can either be autonomous or vector-mediated. Autonomous autogamy is self-

fertilisation in the absence of pollinators whereas vector-mediated autogamy is within flower 

pollen transfer using biotic vectors (Bowers, 1975; Solís-Montero et al., 2021). When 

determining the reproductive strategy used by a species, three major aspects should be 

considered: i) whether there are occurrences of sexual reproduction, ii) whether individuals 

are bisexual or unisexual, and iii) whether the co-sexual individuals are self-compatible 

(Charlesworth, 2006). 

Co-sexuality is the rule in angiosperms (Charlesworth, 2006). The presence of both sexes 

within the same flower (hermaphroditism) can complicate reproduction by promoting self-

pollination (Harder et al., 2000) as well as interfering with sexual function (Lloyd and Yates 

1982; Harder et al., 2000; Dai and Galloway, 2011). The effects of hermaphroditism are 

mediated by reducing pollen-stigma interference (i.e., one sexual function being obstructed by 

another) and promoting outcrossing by ensuring that pollen presentation and stigma receptivity 

occurs non-simultaneously (Ramirez, 2005). This is accomplished by temporally separating 

sex organs (dichogamy) or spatially separating sex organs (herkogamy) (Lloyd and Webb, 



3 
 

1986; Webb and Lloyd, 1986; Harder et al., 2000). Dichogamy may be protogynous (the 

dispersal of pollen after stigma receptivity) or protandrous (the maturation and dispersal of 

pollen before stigma receptivity) (Stout, 1928; Sargent and Otto, 2004). On the other hand, 

herkogamy includes monoecy, dioecy and gynomonoecy. Monoecy is when staminate flowers 

and pistillate flowers are separate but occur on the same plant whereas dioecy is when staminate 

flowers and pistillate flowers occur on different individual plants. Some studies hypothesise 

that these spatial segregations evolved to reduce self-pollination and thereby prevent 

inbreeding depression (Harder et al., 2000); however, Barrett (2003) noted that most plants that 

exhibit herkogamy are already exempted from selfing by physiological mechanisms. Instead, 

herkogamy may play a more important role in reducing sexual interference between maternal 

and paternal functions (Fetscher, 2001; Barrett, 2002). 

Despite most flowering plants being hermaphroditic and having the potential for self-

fertilisation, they usually develop self-incompatibility (SI) systems to enforce outcrossing 

(Goldberg et al., 2010). SI systems enable plants to recognize their own pollen and 

subsequently reject it (Goldberg et al., 2010). This rejection can occur on the stigmatic surface 

(sporophytic incompatibility system) or after the pollen grain has germinated and the pollen 

tube has penetrated the stigma (gametophytic incompatibility system) (Newbigin et al., 1993). 

Gametophytic incompatibility is common in Solanaceae, Rosaceae, and Papaveraceae (e.g., 

Miller and Kostyun, 2011; Del Duca et al., 2019; Bilinski and Kohn, 2012), whereas 

sporophytic incompatibility is often reported in Brassicaceae, Asteraceae, and Convolvulaceae 

(e.g., Mable et al., 2003; Hiscock et al., 2003; Hou et al., 2021). 

SI has a short-term disadvantage when there is no outcross pollen available and reproduction 

is hindered (Newbigin et al., 1993), which can lead to the breakdown of SI systems (Newbigin 

et al., 1993). Transitions from SI to self-compatibility is a common evolutionary shift seen in 

angiosperms and self-compatibility is a prerequisite for self-fertilization (Stebbins, 1974; Igic 



4 
 

et al., 2008). Self-fertilization allows species to reproduce uniparentally in the absence of 

pollinators or in unpredictable environments (Pannell and Barett, 1998; Rea and Nasrallah, 

2010). While self-fertilization provides reproductive assurance, it is regarded as an 

evolutionary ‘dead-end’ (Dobzhansky, 1950; Stebbins, 1957; Igic and Busch, 2013). Selfing 

taxa have a lower potential for adaptation which may be linked to genetic drift and limited 

recombination efficiency (Nöel et al., 2017). Due to increased homozygosity, inbreeding 

depression allows for increased frequencies of recessive traits or mutations to be expressed 

(Charlesworth, 2006). Selfing taxa are therefore more prone to extinction than outcrossing taxa 

(Stebbins, 1957; Glémin et al., 2006; Schoen and Busch, 2008).  

In contrast, outcrossing taxa have high recombination rates and consequently maintain high 

effective population sizes; they exhibit better responses to purifying selection than selfing taxa 

and display high genetic diversity across their geographic distribution (Goldberg et al., 2010). 

Despite the fitness advantages of outcrossing, outbreeding depression can promote a change 

from outcrossing to selfing (Charlesworth, 2006). Outbreeding depression typically occurs 

when there is gene flow between fragmented populations and favourable gene combinations 

are broken up therefore causing an increase in the frequency of maladapted gene complexes 

(Charlesworth, 2006). 

The reproductive strategies used by a plant species have implications for the ecology, 

evolution, and genetic diversity within populations of the species (Charlesworth, 2006). It is 

evident that both outcrossing and selfing have implications for the genetic variation and 

consequently the evolutionary patterns within populations (Richards, 1996). Many species 

adopt a ‘mixed’ reproductive strategy as both outcrossing and selfing confer fitness benefits to 

the offspring (Richards, 1996). Additionally, many perennial plants may be able to reproduce 

asexually, and their evolutionary potential may be linked to the association between 



5 
 

outcrossing and selfing or outcrossing and asexual reproduction (Richards, 1996; Bengtsson 

and Ceplitis, 2000). 

Asexual reproductive strategies  

 

Asexual reproduction includes vegetative propagation and apomixis. Vegetative propagation 

is the formation of progeny by specialized structures, such as bulbs, tubers, and rhizomes, 

rather than seeds or spores (Richards, 1997; Hojsgaard and Hörandl, 2019). Meiosis and 

syngamy are circumvented and offspring have the same genetic composition as the maternal 

plant (Richards, 1997). Clonal populations often occur in environments where sexual 

reproduction is prevented by either a lack of suitable pollinators or unfavourable ecological 

conditions (Barrett, 2015). 

Apomixis is clonal reproduction through unfertilised seeds (Asker and Jerling, 1992). There 

are two types of apomixis: sporophytic (or adventitious embryony) or gametophytic apomixis. 

Sporophytic apomixis occurs when the nucellus or integument of the ovule gives rise to an 

embryo which develops via mitotic division (Koltunow et al., 1995). Sporophytic apomixis is 

the most taxonomically widespread apomictic development pathway, however the genetic 

mechanisms are not as well understood as those of gametophytic apomixis (Hojsgaard and 

Hörandl, 2015; Hojsgaard and Hörandl, 2019). 

Gametophytic apomixis occurs via two distinct mechanisms: diplospory and apospory. 

Diplospory occurs when a megaspore mother cell produces an unreduced embryo sac by 

mitosis or modified meiosis. Apospory occurs when somatic cells, often from the nucellus of 

the ovule, give rise to an unreduced embryo sac (Bertasso-Borges and Coleman, 2005). After 

the production of unreduced embryo sacs in both mechanisms, meiosis is bypassed, and the 

unreduced ova (egg-like cells) develop by parthenogenesis – a form of asexual reproduction 

where the unfertilised egg cells give rise to new individuals (Nogler, 1984). The subsequent 



6 
 

endosperm formation can be pseudogamous or autonomous. In pseudogamy, the polar nuclei 

are fertilised and only the egg cell develops by parthenogenesis. In autonomy, the polar nuclei 

and egg cell are independent of fertilisation, and both develop by parthenogenesis (Nogler, 

1984). 

True obligate apomictic species, that can form seeds exclusively by apomixis, are rare (Asker 

and Jerling, 1992). Although most reports of obligate apomixis were dismissed as it was 

believed to be an ‘artifact of screening tools’ (Asker and Jerling, 1992), some incidences of 

obligate apomixis have been published (e.g., Connor and Dawson, 1993; Grusz et al., 2021). 

Sorensen et al. (2009) provided compelling evidence for the expression of obligate apomixis 

in Corunastylis apostasioides Fitzg, the common midge orchid. Apomictic characteristics in 

the orchid species included seed development without fertilisation, periodically closed flowers 

that still produced mature embryos, subsequent endosperm by diplospory and adventitious 

embryony, inviable pollen grains, and the expansion of ovaries despite the lack of fertilization 

(Sorensen et al., 2009). The expression of apomixis was linked to the inability of C. 

apostasioides to produce a citronella scent to attract pollinators (Sorensen et al., 2009).  

Most apomictic species maintain a degree of sexuality and are referred to as ‘facultative 

apomictics’ (Asker and Jerling, 1992; Richards, 2003). Apomixis and sexuality are thus not 

mutually exclusive reproductive strategies; therefore, it is difficult to predict their role in 

evolution and speciation (Hojsgaard et al., 2014). However, the combination of sexual and 

asexual reproductive pathways within a species is hypothesised to be more beneficial than 

obligate outcrossing (Cosendai et al., 2013). Facultative apomictics benefit from the 

advantages of both sexual and asexual reproduction; asexual reproduction enables apomictic 

populations to rapidly colonize novel niches because they are not limited by mate or pollinator 

availability (Baker, 1965; Baker,1967; Pannell et al., 2015), whereas sexual reproduction 

maintains genetic diversity within the population (Barcaccia et al., 2006).  



7 
 

On the other hand, apomixis and sexual reproductive pathways are mutually exclusive in the 

ovaries of diplosporous species. This is in contrast to aposporous species where both sexual 

reproductive pathways and apomixis can co-exist within the same ovary (Asker and Jerling, 

1992; Hojsgaard et al., 2013). This results in seeds containing either sexually derived embryos 

or parthenogenetic embryos with variable genotype frequencies (Koltunow et al., 1995; 

Hojsgaard et al., 2013). The presence of both reproductive pathways within a plant has a direct 

impact on its genetic contribution to the population’s gene pool (Hojsgaard et al., 2013). 

Offspring derived from facultative aposporous apomixis often have varying levels of genetic 

diversity (Hojsgaard et al., 2013). This genetic variation, however, was presumed to be 

significantly lower than in populations with plants that reproduce predominantly through 

sexual reproductive pathways such as cross-pollination (Asker and Jerling, 1992). 

In contrast, apomictic populations have high proportions of heterozygotes across multiple loci, 

thereby maintaining genetic variation within individuals (Halkett et al., 2005), however, due 

to the lack of recombination, these populations are genotypically depauperate (Peredo,, 2013; 

Grusz and Pryer, 2015). This is not applicable to all apomictic groups because the maintenance 

of male meiosis may generate genotypic variation, thereby providing an evolutionary 

advantage over apomictic groups with no incidences of meiosis (Whitton et al., 2008). 

Additionally, variation within apomictic lineages may arise when asexual individuals continue 

producing sexually functional male gametes and thereby mate with sexual individuals. This 

transmission of apomixis genes to sexually producing plants may contribute to the long-term 

spread of apomixis and generate genotypically diverse populations – even if these events occur 

infrequently (Adolfsson and Bengtsson, 2007; Whitton et al., 2008).  

 

 



8 
 

The co-occurrence of apomixis and polyploidy 

 

Apomixis is often associated with polyploidy (Carman, 1997). The presence of odd ploidy 

levels, such as triploid and pentaploid, is generally a reliable indicator of apomixis – with most 

natural apomictic populations being polyploids (Asker and Jerling, 1992; Koltunow and 

Grossniklaus, 2003). This might be attributed to tetraploid and hexaploid individuals 

maintaining the capability for selfing, which often results in fixed heterozygosity that can 

reduce their potential for establishment (Liu et al., 2012). In contrast, in triploid and pentaploid 

individuals, the transition to apomixis helps overcome female sterility caused by meiotic 

incompatibilities in uneven ploidy levels (Liu et al., 2012). 

Carman (1997) proposed that polyploidy may have triggered apomixis during the Pleistocene 

due to cycles of glaciation and deglaciation in North America and Eurasia. Most apomictic 

families such as Asteraceae, Poaceae, and Orchidaceae, have apomictic species that are 

presumed to have a Pleistocene origin. Repeated deglaciation events resulted in these areas 

being revegetated by species adapted to cool climates and short growing seasons, with 

adaptations for precocious meiosis and embryo sac development. The climatic fluctuations 

would have also caused range shifts that would have resulted in secondary hybrid contact zones 

between high latitude and low latitude flora which facilitated the formation of polyploid 

populations with different cytotypes. Gene flow between cytotypes within these zones may 

have led to asynchronous gene expression in the megasporogenesis and metagametogenesis 

phases of sexual reproduction (e.g., Polegri et al., 2010). This have may resulted in the 

megasporogenesis phase being skipped, leading to the suppression of sexuality and the 

expression of apomixis. This was supported by Grimanelli et al., (2003) who found that the 

frequencies of apomictic phenotypes were influenced by the timing of the steps in 

developmental pathways from meiosis to post-fertilization. 

https://www.frontiersin.org/articles/10.3389/fpls.2019.00358/full#B102


9 
 

The relationship between polyploidy and apomixis is confounded by their co-occurrence, 

making it difficult to discern the causal mechanism of both. Nevertheless, some studies suggest 

that polyploidization may indirectly contribute to the establishment of apomictic individuals 

(e.g., Bierzychudek, 1985; Hojsgaard, 2018; Hojsgaard and Hörandl, 2019), while others 

suggest that apomixis may promote the establishment of polyploid populations (e.g., 

Karunarathne et al., 2018; Kirchheimer et al., 2018). Interestingly, despite the lack of 

knowledge regarding their origins, both polyploidy and apomixis are typically associated with 

invasion potential (Grusz et al., 2009; Pandit et al., 2011; Beck et al., 2011; te Beest et al., 

2012).  

The range expansion in apomictics is often linked to geographical parthenogenesis – a 

phenomenon where asexual organisms occupy a wider distribution, typically at higher altitudes 

or in harsher environments, than their sexual relatives (Vandel, 1928; van Dijk, 2003; Hörandl, 

2009). This might be accounted for using Baker’s Law which states that long distance dispersal 

typically favours self-fertilising and apomictic species because they are not limited by mate or 

pollinator availability (Baker, 1955; Stebbins 1957; Baker, 1967). Additionally, only one 

individual capable of uniparental reproduction is needed to establish an entire population 

(Baker, 1967). For example, a study conducted on Iridaceae in South Africa, found that species 

with higher rates of uniparental reproduction were able to naturalize whereas those without the 

capacity for uniparental reproduction failed to naturalize (van Kleunen and Fischer, 2008). 

Much of the polyploid literature debates the invasion potential of polyploids within the context 

of their ability to occupy new habitats (te Beest et al., 2012; Hahn et al., 2012; Moura et al., 

2021). However, the reproductive strategies of polyploids are probably more important when 

escaping minority cytotype exclusion (MCE; Levin, 1975; Stebbins, 1985). MCE describes a 

situation where the establishment of newly formed polyploids is limited by mate availability 

because they are less common in a randomly mating population. Polyploids were presumed to 

https://link.springer.com/article/10.1007/s00606-020-01692-6#ref-CR8
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6454013/#B59


10 
 

overcome these limitations by occupying habitats outside their diploid parent populations’ 

ecological range thus accounting for the extensive reports of polyploidy in invasive species 

(Hollingsworth and Bailey, 2000; Pandit et al., 2011; te Beest et al., 2012; Baudel et al.,2018). 

However, this was challenged by various studies that noted that niche shifts or the broadening 

of niches does not necessarily contribute to the establishment of polyploid cytotypes 

(Theodoridis et al., 2013; Glennon et al., 2014; Rice et al., 2019). Rather, the success of 

polyploid lineages may be attributed to the competitive advantages provided by asexual 

reproduction (e.g., Kirchheimer et al., 2018). 

Male fertility, polyploidy and apomixis 

 

Male fertility is an important factor to consider when investigating invasive species. High male 

fertility could facilitate interploidy gene flow which may generate new beneficial traits in fertile 

progenies (Alexander, 2020). Assessing the fertility of potential parents may provide insight 

on cross combinations that may occur within a species (Alexander, 2020). Male fertility in 

polyploid species has implications for the reproductive strategies used in populations (Atlagić 

et al., 2012) and the subsequent genetic variation within these populations (Whitton et al., 

2008). 

Ramsey and Schemske (1998) noted that pollen grains of variable ploidy level can facilitate 

the formation of polyploid or hybrid cytotypes – this may have long term implications for the 

invasion of novel niches, establishment of populations and an increased distribution of 

genotypes. For example, interploidy mating between triploids and tetraploids produces 

polyploid progeny with odd chromosome numbers (Peckert and Chrtek, 2006). This results in 

the progeny having novel combinations of the parental genes thereby increasing the gene pool 

in populations which may provide an adaptive advantage in invaded regions (Rotreklová and 

Krahulcová, 2006; Alexander 2020).  



11 
 

In polyploid agamic complexes (i.e., sexual reproduction is partly or completely replaced by 

asexual reproduction), high pollen grain viability may be linked to the expression of apomixis 

(Rotreklová and Krahulcová, 2006).  In both facultative apomictics and pseudogamous 

apomictic species, endosperm development is still dependent on fertilisation; therefore, the 

production of viable pollen is maintained in most apomictics (Nogler, 1984; Asker and Jerling, 

1992). This allows for genes to be transmitted to facultative apomictic and sexual relatives, 

which enables the formation of genetically diverse clone complexes (Whitton et al., 2008; van 

Dijk, 2009).  

The presence of clonal diversity within a population makes biocontrol using a genotype-

specific agent challenging (van Dijk, 2003; Chapman et al., 2004). A study conducted on 

invasive Chondrilla juncea L. (skeleton weed) in Australia found that the introduced biocontrol 

agent, the rust fungus Puccinia chondrillina, had different effects on the survival rate of the 

three clones within the population (Burdon et al., 1981). The frequency of one clone type 

increased while the other two decreased. Conversely, in the native range of Chondrilla juncea 

in Turkey, the most abundant clones were over-infected (Burdon et al., 1981). Investigating 

the sources of genetic diversity, such as reproductive strategies and polyploidy, within invasive 

taxa is therefore important before releasing biocontrol agents.  

Campuloclinium macrocephalum 

 

Campuloclinium macrocephalum (Asteraceae, Eupatorieae), or ‘pompom weed’, is endemic to 

South America – specifically Brazil, Paraguay, Uruguay, Bolivia, Argentina, Central America 

Farco et al., 2012). The species has medicinal value in these areas and the leaves are often used 

as a sedative and anti-inflammatory (Vega et al., 2008). C. macrocephalum was introduced to 

South Africa in the 1960s and the earliest population was found in Pretoria (McConnachie et 

al., 2011). By the 1980s the species had spread rapidly across Gauteng and into Limpopo 



12 
 

(Henderson, 2007). The species then entered an exponential expansion phase and its range 

increased to include Mpumalanga, Limpopo, and North-West provinces (Henderson, 2007) 

(Figure 1.1, this thesis). 

The pompom weed is a perennial herb that develops from a woody rootstock with perennating 

buds. The bright pink capitula (Figure 1.2) comprise multiple florets with ovaries that mature 

into cypselas (asteraceous fruits with a single seed). The cypselas have a persistent pappus 

(Goodall et al., 2010). Small clusters of capitula on a common peduncle form synflorescences. 

The plant invests a considerable amount of its resources in underground structures to ensure its 

survival during winter months (Goodall et al., 2010). Moisture availability is the main 

environmental stressor for the pompom weed as the onset of winter causes the plants to senesce 

(Goodall et al., 2010). The weed is typically found in areas that receive more than 600 mm of 

rain per annum, however the amount of rainfall that limits establishment and spread is unknown 

(Goodall et al.,2010). 

Campuloclinium macrocephalum typically occupies disturbed sites, such as road margins and 

old fields and is consequently considered a pioneer species. The species thereafter invades 

grasslands, savannas, and wetlands (McConnachie et al., 2011) and transforms such 

ecosystems to the extent that the indigenous species are eliminated. The species does this by 

increasing soil erosion, reducing biodiversity by displacing native species, interfering with the 

establishment of grass species, and reducing the carrying capacity of invaded areas (Dixon, 

2008). Due to the absence of natural enemies in the non-native region/country, this species’ 

distribution was predicted to exponentially increase every year unless a successful biocontrol 

agent was found (Trethowan et al., 2011).  

Invasion opportunity into rangelands is predominantly through wind-dispersal; however, 

establishment is restricted by vegetation quality – specifically sward and basal cover (Goodall 



13 
 

et al., 2010). The vegetation invaded by this weed is usually in poor condition and typically 

occurs in rangelands affected by unsustainable grazing, poorly managed/abandoned 

agricultural fields and drained wetlands (Goodall et al., 2010). Campuloclinium 

macrocephalum does not invade undisturbed grasslands (Lake and Leishman, 2004; Goodall 

et al., 2010). This species reduces the grazing potential of grasslands as it is unpalatable to 

livestock. This may be due to the glandular trichomes on stems and leaves producing 

phytotoxic substances (McConnachie et al., 2011). The success of pompom weed’s spread is 

not attributed to its allelopathic nature (McConnachie et al., 2011), but could be due to its 

ability to reproduce prolifically and asexually.  

The eradication of C. macrocephalum has proven to be a difficult endeavour (McConnachie et 

al., 2011). In the past, management plans for the pompom weed have relied on mechanical and 

chemical methods. The mechanical methods, such as hoeing and ploughing, were unsuccessful 

as it encouraged regeneration from the xylopodium and the plants subsequently thrived in the 

disturbed environments (McConnachie et al., 2011). Chemical methods, such as herbicides, 

negatively impacted other vegetation in the vicinity and use of this method is restricted to 

roadsides (McConnachie et al., 2011). Additionally, C. macrocephalum does not use vegetative 

reproduction, however the xylopodium of the woody underground rootstock functions as a 

storage organ (Farco et al., 2012). This may be co-opted for persistence of the plant if the above 

ground portion dies (Farco et al., 2012). Therefore, the pompom weed is resistant to herbicide 

damage, fire, and drought, resulting in it being a successful invader (Farco et al., 2012).  

Biological control programmes using Liothrips tractabilis Uzel 1978 (thrips) and Puccina 

eupatorii Cummins 1978 (leaf rust pathogen) have produced better results than both the 

mechanical and chemical methods. The thrips cause foliage deformation thereby reducing 

growth and biomass accumulation – particularly in younger plants (Ramanand et al., 2016; 

Ramanand et al., 2017). The leaf rust pathogen causes premature senescence; however, this 



14 
 

stimulates the production of compensatory growth in late autumn (Goodall et al., 2012). 

Therefore, the leaf rust pathogen is unlikely to cause measurable reductions in pompom weed 

densities (Goodall et al., 2012).  

The success of biocontrol agents on a species may be linked to the type of reproductive system 

used by a species, polyploidy, and the consequent genetic variation within a species. A previous 

study conducted on the reproductive strategy of the pompom weed found that the species is 

capable of uniparental reproduction, however it was unclear if it was autonomous autogamy or 

apomixis (Kgaboesele, unpubl. data). Gitonga et al. (2015) found low genetic variability 

amongst populations of the pompom weed, but the presence of apomixis could not be 

confirmed. Gitonga et al., (2022) identified triploid and tetraploid cytotypes of C. 

macrocephalum in South African populations, however the predominant reproductive strategy 

used by populations of C. macrocephalum in South Africa remains unclear. 

Aim and objectives  

 

The aim of this investigation was to infer whether populations of Campuloclinium 

macrocephalum (pompom weed) reproduce by vector-mediated crosses, self-pollination or 

apomixis (either facultative or obligate) and assess the relationship between polyploidy and the 

mode of reproduction.  

The objectives associated with this aim were to:  

1) Survey insect visitors to capitula and investigate whether effective vector-mediated 

cross-pollination may occur in populations of C. macrocephalum in Gauteng. 

2) Compare pollen size, shape, and viability between selected triploid and tetraploid 

populations to assess male fertility and infer possible reproductive barriers in 

populations of different ploidy levels. 



15 
 

3) Assess whether pollen tubes are reaching the ovaries and/or ovules in bagged capitula 

to determine if autogamy or apomixis is predominantly occurring, and the type of 

apomixis.  

4) Compare genetic diversity within and among triploid and tetraploid populations of C. 

macrocephalum in order to infer whether plants self-fertilise or reproduce via apomixis, 

and if apomixis is occurring – assess whether it is facultative or obligate. 

 Dissertation outline 

  

This thesis is divided into four chapters: the current introductory chapter, chapter 2, chapter 3, 

and a concluding chapter. Due to the structure of the thesis, there may be some repetition in 

background material presented. The chapters are listed below: 

1) An introductory chapter that comprises the rationale and literature review, aim and 

objectives of the study. 

2) A characterisation of male fertility in triploid and tetraploid populations of C. 

macrocephalum. 

3) Reproductive biology of C. macrocephalum and its applicability in biocontrol management 

plants. 

4) A concluding chapter to synthesise the thesis results and recommendations for future studies 



16 
 

 

Figure 1.1. The current recorded distribution of Campuloclinium macrocephalum in the 

Gauteng (G), Limpopo (L), Mpumalanga (M), Kwa-Zulu Natal (K) and Western Cape (W) 

provinces of South Africa. The location data were obtained from the SANBI NewPOSA 

database (Last updated: 14 October 2021). 

 

 



17 
 

 

Figure 1.2. Dense populations of Campuloclinium macrocephalum in Greenstone (Gauteng), 

characterized by bright pink florets, during the December 2020 flowering period.  



18 
 

REFERENCES  

Abrol, D. (2012). Honeybee and crop pollination. In: Abrol, D. (ed.) Pollination biology. 

Springer, Dordrecht. 

Adolfsson, S. and Bengtsson, B. (2007). The spread of apomixis and its effect on resident 

genetic variation. Journal of Evolutionary Biology, 20: 1933–1940. 

Aileen, J. (2005). The impact of the invasive plant, Campuloclinium macrocephalum (Less.) 

DC., on plant community structure in the Rietvlei Nature Reserve, Pretoria. Honours 

Research report GGY 702. Department of Geography, Geoinformatics and 

Meteorology, University of Pretoria, Pretoria, South Africa. 

Alexander, L. (2020). Ploidy level influences pollen tube growth and seed viability in 

interploidy crosses of Hydrangea macrophylla. Frontiers in Plant Science, 2020: 100. 

Asker, S. and Jerling, L. (1992). Apomixis in plants. CRC Press, Florida. 

Atlagić, J., Terzic., S., Marjanović -Jeromela, A. (2012). Staining and fluorescent microscopy 

methods for pollen viability determination in sunflower and other plant 

species. Industrial Crops and Products, 35: 88–91.  

Baker, H. (1955). Self-compatibility and establishment after long-distance 

dispersal. Evolution, 9: 347–349. 

Baker, H. (1965). Characteristics and modes of origin of weeds. The genetics of colonizing 

species., 147–168. 

Baker, H. (1967). Support for Baker's law-as a rule. Evolution, 21: 853–856. 



19 
 

Barcaccia, G., Arzenton, F., Sharbel, T., Varotto, S., Parrini, P. and Lucchin, M. (2006). 

Genetic diversity and reproductive biology in ecotypes of the facultative apomict 

Hypericum perforatum L. Heredity, 96: 322–334. 

Barrett, S. (2002). Sexual interference of the floral kind. Heredity, 88: 154–159. 

Barrett, S. (2003). Mating strategies in flowering plants: the outcrossing–selfing paradigm and 

beyond. Philosophical Transactions of the Royal Society of London. Series B: 

Biological Sciences, 358: 991–1004.  

Barrett, S. (2015). Influences of clonality on plant sexual reproduction. Proceedings of the 

National Academy of Sciences, 112: 8859–8866. 

Baudel, P., Bray, S., Vallejo-Marin, M., Kolar, F. and Yant, L. (2018). The “Polyploid Hop”: 

Shifting challenges and opportunities over the evolutionary lifespan of genome 

duplications. Frontiers in Ecology and Evolution, 6: 117–121. 

Beck, J., Windham, M. and Pryer, K. (2011). Do asexual polyploid lineages led short lives? A 

case study from the fern genus Astrolepis. Evolution, 65: 3217–3229. 

Bengtsson, B. and Ceplitis, A. (2000). The balance between sexual and asexual reproduction 

in plants living in variable environments. Journal of Evolutionary Biology, 13: 415–

422. 

Bertasso-Borges, M. and Coleman, J. (2005). Cytogenetics and embryology of Eupatorium 

laevigatum (Compositae). Genetics and Molecular Biology, 28: 123–128. 

Bierzychudek, P. (1985). Patterns in plant parthenogenesis. Experientia, 41: 1255–1264. 

Bilinski, P. and Kohn, J. (2012). Sites of self-pollen tube inhibition in Papaveraceae (sensu 

lato). Plant Systematics and Evolution, 298: 1239–1247. 



20 
 

Bowers, K. (1975). The pollination ecology of Solanum rostratum (Solanaceae). American 

Journal of Botany, 62: 633–638. 

Burdon, J., Groves, R. and Cullen, J. (1981). The impact of biological control on the 

distribution and abundance of Chondrilla juncea in south-eastern Australia. Journal of 

Applied Ecology, 957–966. 

Carman, J. (1997). Asynchronous expression of duplicate genes in angiosperms may cause 

apomixis, bispory, tetraspory and polyembrology. Biological Journal of the Linnean 

Society, 61: 51–94. 

Chapman, H., Robson, B. and Pearson, M. (2004). Population genetic structure of a colonising, 

triploid weed, Hieracium lepidulum. Heredity, 92: 182–188. 

Charlesworth, D. (2006). Evolution of plant breeding systems. Current Biology, 16: 726–735. 

Connor, H. and M. Dawson (2003). Evolution of reproduction in Lamprothyrsus (Arundineae: 

Gramineae). Annals of the Missouri Botanical Garden, 1993: 512–517. 

Cosendai, A., Wagner, J., Ladinig, U., Rosche, C. and Hörandl, E. (2013). Geographical 

parthenogenesis and population genetic structure in the alpine species Ranunculus 

kuepferi (Ranunculaceae). Heredity, 110: 560–569. 

Dai, C. and Galloway, L. (2011). Do dichogamy and herkogamy reduce sexual interference in 

a self‐incompatible species? Functional Ecology, 25: 271–278. 

Del Duca, S., Aloisi, I., Parrotta, L. and Cai, G. (2019). Cytoskeleton, transglutaminase and 

gametophytic self-incompatibility in the Malinae (Rosaceae). International journal of 

molecular sciences, 20: 209. 

Dijk, P., Jong, H., Vijverberg, K. and Biere, A. (2009). An apomixis-gene’s view on 

dandelions. In Schöen, I., Martens, K. and van Dijk, P Lost Sex. Springer, Dordrecht. 



21 
 

Dixon, G. (2008). Allelopathic potential of the alien invader weed Campuloclinium 

macrocephalum (Less) DC. Agronomy thesis, Department of Plant, Production and 

Soil Science, Faculty of Natural and Agricultural Sciences, University of Pretoria.  

Dobzhansky, T. (1950). Evolution in the tropics. American scientist, 38: 209–221. 

Farco, G., Sosa, M., Dematteis, M. and Fernández, A. (2012). Cytology and embryology of the 

pompom weed, Campuloclinium macrocephalum (Eupatorieae, Asteraceae). South 

African Journal of Botany, 78: 21–29. 

Fetscher, A. (2001). Resolution of male-female conflict in a hermaphroditic 

flower. Proceedings of the Royal Society of London. Series B: Biological Sciences, 268: 

525–529. 

Gitonga, L., Cron, G., Glennon, K., McConnachie, A., and Byrne, M. (2022). Two ploidy levels 

present in the invasive Campuloclinium macrocephalum (pompom weed) in South 

Africa–Implications for biocontrol. Weed Research, 62: 59–67. 

Gitonga, L., Cron, G., McConnachie, A., Glennon, K. and Bryne, M. (2015). Genetic variation 

of two invasive Campuloclinium macrocephalum, Asteraceae in South Africa, inferred 

from molecular markers. Weed Research, 55: 51–61. 

Glémin, S., Bazin, E. and Charlesworth, D. (2006). Impact of mating systems on patterns of 

sequence polymorphism in flowering plants. Proceedings of the Royal Society B: 

Biological Sciences, 273: 3011–3019. 

Glennon, K., Ritchie, M. and Segraves, K. (2014). Evidence for shared broad‐scale climatic 

niches of diploid and polyploid plants. Ecology letters, 17: 574–582. 

Goldberg, E., Kohn, J., Lande, R., Robertson, K., Smith, S. and Igić, B. (2010). Species 

selection maintains self-incompatibility. Science, 6003: 493–495. 



22 
 

Goodall, J., Witkowski, E., Amman, S. and Reinhardt, C. (2010). Does allelopathy explain the 

invasiveness of Campuloclinium macrocephalum (pompom weed) in the South African 

grassland biome? Biological Invasions, 12: 3497–3512. 

Goodall, J., Witkowski, E., McConnachie, A. and Keen, C. (2012). Altered growth, population 

structure and realised niche of the weed Campuloclinium macrocephalum (Asteraceae) 

after exposure to the naturalised rust Puccinia eupatorii (Pucciniaceae). Biological 

Invasions, 14: 1947–1962. 

Grimanelli, D., García, M., Kaszas, E., Perotti, E. and Leblanc, O. (2003). Heterochronic 

expression of sexual reproductive programs during apomictic development in 

Tripsacum. Genetics, 165: 1521–1531. 

Grusz, A. and Pryer, K. (2015). Development of microsatellite markers for the apomictic 

triploid fern Myriopteris lindheimeri (Pteridaceae). Applications in Plant Sciences, 3: 

1500061. 

Grusz, A., Windham, M, Picard, K., Pryer, K., Schuettpelz, E. and Haufler, C. (2021). A 

drought‐driven model for the evolution of obligate apomixis in ferns: evidence from 

pellaeids (Pteridaceae). American Journal of Botany, 108: 263–283. 

Grusz, A., Windham, M. and Pryer, K. (2009). Deciphering the origins of apomictic polyploids 

in the Cheilanthes yavapensis complex (Pteridaceae). American Journal of Botany, 96: 

1636–1645. 

Hahn, M, Van Kleunen, M. and Müller-Schärer, H. (2012). Increased phenotypic plasticity to 

climate may have boosted the invasion success of polyploid Centaurea stoebe. PloS 

one, 7: e50284. 



23 
 

Halkett, F., Simon, J. and Balloux, F. (2005). Tackling the population genetics of clonal and 

partially clonal organisms. Trends in Ecology and Evolution, 20: 194–201. 

Harder, L., Barrett, S. and Cole, W. (2000). The mating consequences of sexual segregation 

within inflorescences of flowering plants. Proceedings of the Royal Society of London. 

Series B: Biological Sciences, 1441: 315–320. 

Hawkes, C. (2007). Are invaders moving targets? The generality and persistence of advantages 

in size, reproduction, and enemy release in invasive plant species with time since 

introduction. The American Naturalist, 170: 832–843. 

Henderson, L. (2001). Alien weeds and invasive plants. Plant Protection Research Institute 

Handbook No. 12, Agricultural Research Council, Pretoria. 

Henderson, L. (2007). Invasive, naturalized and casual alien plants in southern Africa: a 

summary based on the Southern African Plant Invaders Atlas (SAPIA). Bothalia, 37:  

215–248. 

Hiscock, S., McInnis, S., Tabah, D., Henderson, C., and Brennan, A. (2003). Sporophytic self‐

incompatibility in Senecio squalidus L. (Asteraceae)—the search for S. Journal of 

Experimental Botany, 54: 169–174. 

Hojsgaard, D. (2018). Transient activation of apomixis in sexual neotriploids may retain 

genomically altered states and enhance polyploid establishment. Frontiers in Plant 

Science, 9: 230. 

Hojsgaard, D. and Hörandl, E. (2015). Apomixis as a facilitator of range expansion and 

diversification in plants. In Pontarotti, P. (ed.) Evolutionary Biology: Biodiversification 

from genotype to phenotype. Springer International Publishing, Switzerland. 



24 
 

Hojsgaard, D. and Hörandl, E. (2019). The rise of apomixis in natural plant plants. Frontiers 

of Plant Science, 10: 358–362.  

Hojsgaard, D., Greilhuber, J., Pellino, M., Paun, O., Sharbel, T. and Hörandl, E. (2014). 

Emergence of apospory and bypass of meiosis via apomixis after sexual hybridisation 

and polyploidisation. New Phytologist, 204: 1000–1012. 

Hojsgaard, D., Martinez, E. and Quarin, C. (2013). Competition between meiotic and apomictic 

pathways during ovule and seed development results in clonality. New Phytologist, 197: 

336–347. 

Hollingsworth, M., and Bailey, J. (2000). Evidence for massive clonal growth in the invasive 

weed Fallopia japonica (Japanese Knotweed). The Botanical Journal of the Linnean 

Society, 133: 463–472.  

Holzmueller, E. J. and Jose, S. (2009). Invasive plant conundrum: What makes the aliens so 

successful? Journal of Tropical Agriculture, 47: 18–29. 

Hörandl, E. (2006). The complex causality of geographical parthenogenesis. New 

Phytologist, 171: 525–538. 

Hou, S., Zhao, T., Yang, D., Li, Q., Liang, L., Wang, G. and Ma, Q. (2021). Selection and 

Validation of Reference Genes for Quantitative RT-PCR Analysis in Corylus 

heterophylla Fisch.× Corylus avellana L. Plants, 10: 159. 

Igic, B. and Busch, J. (2013). Is self‐fertilization an evolutionary dead end? New 

Phytologist, 198: 386–397. 

Igic, B., Lande, R. and Kohn, J. (2008). Loss of self-incompatibility and its evolutionary 

consequences. International Journal of Plant Sciences, 169: 93–104. 



25 
 

Karunarathne P., Schedler M., Martinez E., Honfi A., Novichkova A. and Hojsgaard D. 

(2018). Intraspecific ecological niche divergence and reproductive shifts foster 

cytotype displacement and provide ecological opportunity to polyploids. Annals of 

Botany, 121: 1183–1196. 

Kgaboesele, P. (unpubl. data). Apomixis and the mode of reproduction in the invasive weed 

Campuloclinium macrocephalum in South Africa. Honours Research Report. 

University of the Witwatersrand, Johannesburg, South Africa.  

Kirchheimer, B., Wessely, J., Gattringer, A., Hülber, K., Moser, D., Schinkel, C. and Dullinger, 

S. (2018). Reconstructing geographical parthenogenesis: Effects of niche 

differentiation and reproductive mode on Holocene range expansion of an alpine 

plant. Ecology Letters, 21: 392–401. 

Koltunow, A. and Grossniklaus, U. (2003). Apomixis: a developmental perspective. Annual 

review of plant biology, 54: 547–574. 

Koltunow, A., Bicknell, R. and Chaudhury, A. (1995). Apomixis: molecular strategies for the 

generation of genetically identical seeds without fertilization. Plant Physiology, 108: 

1345–1352. 

Krahulcová, A. and Krahulec, F. (2021). Cytotype variation and clonal diversity in polyploid 

apomictic populations of Pilosella (Compositae Cichorieae) introduced to southern 

Patagonia. Boletin de la Sociedad Argentina de Botanica, 56: 307–326. 

Lake, J. and Leishman, M. (2004). Invasion success of exotic plants in natural ecosystems: the 

role of disturbance, plant attributes and freedom from herbivores. Biology 

Conservation, 117: 215–226. 

Levin, D. A. (1975). Minority cytotype exclusion in local plant populations. Taxon, 24: 35–43. 



26 
 

Liu, H., Dyer, R., Guo, Z., Meng, Z., Li, J. and Schneider, H. (2012). The evolutionary 

dynamics of apomixis in ferns: a case study from polystichoid ferns. Journal of 

Botany, 2012. 

Lloyd, D. and Webb, C. (1986). The avoidance of interference between the presentation of 

pollen and stigmas in angiosperms I. Dichogamy. New Zealand Journal of Botany, 24: 

135–162. 

Lloyd, D. and Yates, J. (1982). Intrasexual selection and the segregation of pollen and stigmas 

in hermaphrodite plants, exemplified by Wahlenbergia albomarginata 

(Campanulaceae). Evolution, 903–913. 

Mable, B., Schierup, M. and Charlesworth, D. (2003). Estimating the number, frequency, and 

dominance of S-alleles in a natural population of Arabidopsis lyrata (Brassicaceae) 

with sporophytic control of self-incompatibility. Heredity, 90: 422–431. 

McConnachie, A., Retief, E., Henderson, L. and Kay, F. (2011). The initiation of a biological 

control programme against Pompom weed, Campuloclinium macrocephalum (Less.) 

DC. (Asteraceae), in South Africa. African Entomology, 19: 258–268. 

Miller, J. and Kostyun, J. (2011). Functional gametophytic self-incompatibility in a peripheral 

population of Solanum peruvianum (Solanaceae). Heredity, 107: 30–39. 

Moura, R., Queiroga, D., Vilela, E. and Moraes, A. (2021). Polyploidy and high environmental 

tolerance increase the invasive success of plants. Journal of plant research, 134: 105–

114. 

Newbigin, E., Anderson, M. and Clarke, A. (1993). Gametophytic self-incompatibility 

systems. The Plant Cell, 5: 1315. 



27 
 

Noël, E., Jarne, P., Glémin, S., MacKenzie, A., Segard, A., Sarda, V. and David, P. (2017). 

Experimental evidence for the negative effects of self-fertilization on the adaptive 

potential of populations. Current Biology, 27: 237–242. 

Nogler G. (1984). Gametophytic Apomixis. In Johri B. (ed.) Embryology of Angiosperms. 

Springer, Berlin. 

Pandit, M., Pocock, M. and Kunin, W. (2011).  Ploidy influences rarity and invasiveness in 

plants. Journal of Ecology, 99: 1108–1115. 

Pannell, J. and Barrett, S. (1998). Baker's law revisited: reproductive assurance in a 

metapopulation. Evolution, 52: 657–668. 

Pannell, J., Auld, J., Brandvain, Y., Burd, M., Busch, J., Cheptou, P. and Winn, A. (2015). The 

scope of Baker's law. New Phytologist, 208: 656–667. 

Peckert, T. and Chrtek, J. (2006). Mating interactions between coexisting dipoloid, triploid and 

tetraploid cytotypes of Hieracium Echioides (Asteraceae). Folia Geobotanica, 41: 

323–334. 

Peredo, E. (2013). Mating system in Blechnum spicant and Dryopteris affinis ssp. affinis 

correlates with genetic variability. American Fern Journal, 103: 27–39. 

Pimentel, D., McNair, S., Janecka, J., Wightman, J., Simmonds, C., O’Connell, C., and 

Tsomondo, T. (2001). Economic and environmental threats of alien plant, animal, and 

microbe invasions. Agriculture, Ecosystems and Environment, 84: 1–20. 

Polegri, L., Calderini, O., Arcioni, S. and Pupilli, F. (2010). Specific expression of apomixis-

linked alleles revealed by comparative transcriptomic analysis of sexual and apomictic 

Paspalum simplex Morong flowers. Journal of experimental botany, 61: 1869–1883. 



28 
 

Ramandand, H., McConnachie, A. and Olckers, T. (2017). Thermal tolerance of Liothrips 

tractabilis, a biological control agent of Campuloclinium macrocephalum recently 

established in South Africa. Entomologia Experimentalis et Applicata, 162: 234–242. 

Ramandand, H., Olckers, T. and McConnachie, A. (2016). Response of the invasive pompom 

weed, Campuloclinium macrocephalum (Asteraceae), to feeding by the foliage-

deforming thrips Liothrips tractabilis (Phlaeothripidae) under outdoor conditions in 

South Africa. Biocontrol Science and Technology, 26: 1643–1651. 

Ramírez, N. (2005). Plant sexual systems, dichogamy, and herkogamy in the Venezuelan 

Central Plain. Flora-Morphology, Distribution, Functional Ecology of Plants, 200: 30–

48. 

Ramsey, J. and Schemske, D. (1998). Pathways, mechanisms, and rates of polyploid formation 

in flowering plants. Annual review of ecology and systematics, 29: 467–501. 

Rea, A., Liu, P., and Nasrallah, J. (2010). A transgenic self-incompatible Arabidopsis thaliana 

model for evolutionary and mechanistic studies of crucifer self-

incompatibility. Journal of Experimental Botany, 61: 1897–1906. 

Retief, E., van Rooi, C. and Den Breeyen, A. (2016). Environmental requirements and host-

specificity of Puccinia eupatorii, a potential biocontrol agent of Campuloclinium 

macrocephalum in South Africa. Australasian Plant Pathology, 45: 135–144. 

Rice, A., Šmarda, P., Novosolov, M., Drori, M., Glick, L., Sabath, N. and Mayrose, I. (2019). 

The global biogeography of polyploid plants. Nature Ecology and Evolution, 3: 265–

273. 

Richards, A. (1996). Genetic variability in obligate apomicts of the genus Taraxacum. Folia 

Geobotanica and Phytotaxonomica, 31: 405–414. 



29 
 

Richards, A. (1997). In Richards, A. (ed.) Plant breeding systems. London: Chapman and Hall. 

Richards, A. (2003). Apomixis in flowering plants: an overview. Philosophical Transactions 

of the Royal Society of London. Series B: Biological Sciences, 358: 1085–1093. 

Rotreklová, O., and Krahulcová, A. (2016). Estimating paternal efficiency in an agamic 

polyploid complex: pollen stainability and variation in pollen size related to 

reproduction mode, ploidy level and hybridogenous origin in Pilosella 

(Asteraceae). Folia Geobotanica, 51: 175–186. 

Sargent, R. and Otto, S. (2004). A phylogenetic analysis of pollination mode and the evolution 

of dichogamy in angiosperms. Evolutionary Ecology Research, 6: 1183–1199. 

Schoen, D. and Busch, J. (2008). On the evolution of self-fertilization in a 

metapopulation. International Journal of Plant Sciences, 169: 119–127. 

Solís-Montero, L., Aceves-Chong, L., Vega-Polanco, M. and Vargas-Ponce, O. (2021). 

Changes in reproductive traits in Physalis philadelphica; an unexpected shift toward 

self-incompatibility in a domesticated annual fruit crop. Frontiers in Plant Science, 12: 

834. 

Sorensen, A., Rouse, D., Clements, M., John, P. and Perotti, E. (2009). Description of a 

fertilization-independent obligate apomictic species: Corunastylis apostasioides 

Fitzg. Sexual Plant Reproduction, 22: 53–165. 

Stebbins, G. (1974). Flowering Plants: Evolution above the species level. Belknap Press, 

Cambridge, Massachusetts. 

Stebbins, G. (1985). Polyploidy, hybridization, and the invasion of new habitats. Annals of the 

Missouri Botanical Garden, 824–832. 



30 
 

Stebbins, G. (1957). Self-fertilization and population variability in the higher plants. The 

American Naturalist, 91: 337-354. 

Stout, A. (1928). Dichogamy in flowering plants. Bulletin of the Torrey Botanical Club, 141–

153. 

Te Beest, M., Le Roux, J., Richardson, D., Brysting, A., Suda, J., Kubešová, M., and Pyšek, P. 

(2012). The more the better? The role of polyploidy in facilitating plant 

invasions. Annals of Botany, 109: 19–45. 

Theodoridis, S., Randin, C., Broennimann, O., Patsiou, T. and Conti, E. (2013). Divergent and 

narrower climatic niches characterize polyploid species of European primroses in 

Primula sect. Aleuritia. Journal of Biogeography, 40: 1278–1289. 

Trethowan, P., Robertson, M. and McConnachie, A. (2011). Ecological niche modelling of an 

invasive alien plant and its potential biological control agents. South African Journal of 

Botany, 77: 137–146. 

Van Dijk, P. (2003). Ecological and evolutionary opportunities of apomixis: insights from 

Taraxacum and Chondrilla. Philosophical Transactions of the Royal Society of London. 

Series B: Biological Sciences, 358: 1113–1121. 

van Kleunen, M. and Fischer, M. (2008). Adaptive rather than non-adaptive evolution of 

Mimulus guttatus in its invasive range. Basic and Applied Ecology, 9: 213–223. 

Vandel, A. (1928). Geographical parthenogenesis: contribution to the biological and 

cytological study of natural parthenogenesis. Biology Bulletin of France and Belgium, 

62: 164–281.  



31 
 

Vega, M., Carvalho, M., Vieira, I. and Braz-Filho, R. (2008). Chemical constituents from the 

Paraguayan medicinal plant, Eupatorium macrocephalum Less. Journal of Natural 

Medicine, 62: 122–123. 

Webb, C. and Lloyd, D. (1986). The avoidance of interference between the presentation of 

pollen and stigmas in angiosperms II. Herkogamy. New Zealand journal of botany, 24: 

163–178. 

Whitton, J., Sears, C., Baack, E. J. and Otto, S.  (2008). The dynamic nature of apomixis in the 

angiosperms. International Journal of Plant Sciences, 169: 169–182. 

Yan, X., Zhou, B., Yin, Z., Wang, N. and Zhang, Z. (2016). Reproductive biological 

characteristics potentially contributed to invasiveness in an alien invasive plant Bidens 

frondosa. Plant Species Biology, 31: 107–116. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



32 
 

Chapter 2 
 

Male fertility and potential gene flow of triploid and tetraploid populations of 

Campuloclinium macrocephalum. 

ABSTRACT 

 

Campuloclinium macrocephalum, commonly referred to as the pompom weed, is an alien 

invasive species in South Africa. The species is currently threatening the survival of biomes in 

South Africa by causing a reduction in indigenous vegetation. Without intervention, the species 

will invade the entire grassland biome and threaten food security. Previous studies 

hypothesised that a species’ invasiveness is linked to ploidy and apomictic reproduction within 

its invaded range. This study assesses male fertility in two triploid and two tetraploid 

populations of C. macrocephalum in Johannesburg, South Africa to test if gene flow is 

occurring amongst populations of different ploidy levels. Pollen from triploid and tetraploid 

populations were stained with aniline blue and acetocarmine; populations exhibited mean 

pollen viability percentages of 98% and 90% which indicates high male viability. Such viability 

could enable gene flow among populations and result in the formation of genetically diverse 

progeny via a ‘triploid bridge’. In contrast, fluorescence microscopy and scanning electron 

microscopy of pollen tubes showed that pollen tube growth is being arrested on the stigmatic 

margin and on the sides of the style branches, which may be indicative of a triploid block. 

Previous studies suggest that triploid blocks can be overcome by the production of viable 

gametes; therefore interploidy mating may be occurring at low frequencies. Interploidy mating 

may result in the production of offspring that exhibit differential responses to biocontrol agents.  

Keywords: Male fertility; pollen viability; polyploidy; interploidy mating; gene flow 



33 
 

 

INTRODUCTION 

 

Ploidy levels have been shown to influence pollen biology within a species (Johansen and 

Bothmer, 1994; Katsiotis and Forsberg, 1995; Zlesak, 2008; Knight et al., 2010). Ploidy can 

often lead to changes in the breeding system strategies used by a plant species therefore a better 

understanding of male fertility is required (Maia et al., 2015). Pollen grains are responsible for 

the transfer of male genetic material and directly contribute to the reproductive output in plants 

(Mert, 2009). Pollen grain characterisation and pollen viability can offer insight into the 

differences in male fertility among populations with different ploidy levels (Johansen and 

Bothmer, 1994; Czarnecki et al., 2014).  

Pollen grain characterisation includes assessing the morphological variation of pollen grains 

within a species – particularly pollen grain size and shape. In many genera, pollen size is 

positively correlated with genome size (Beaulieu et al., 2008; Knight et al., 2010) and ploidy 

level (Katsiotis and Forsberg, 1995; Zlesak, 2008). There is also a trade-off between pollen 

grain size and the amount of pollen produced (e.g., Vonhof and Harder, 1995). Selfing species 

often produce fewer pollen grains than outcrossing species, resulting in selfing species having 

larger pollen grains (Cruden, 2000; Naghiloo and Siahkolaee, 2019). Additional pollen grain 

characteristics such as apertures, projections on the pollen grain surface, and clumping has 

implications for the pollination mechanism and consequent reproductive strategy used by a 

species (Mignot et al., 1994, Nagihloo et al., 2018 and Ackerman, 2000). These factors may 

influence pollen viability (Till-Bottraud et al., 1994), increase the efficiency of pollinator 

interactions (Naghiloo et al., 2018) or facilitate pollen dispersal (Ackerman, 2000). 

Pollen grain morphology is also influenced by polyploidy (Liu et al., 2003). Tetraploids 

produce larger pollen grains than both diploids and triploids (Tas and van Dijk, 1999). Larger 

pollen grains in diploids increase reproductive success as there is a strong relationship between 



34 
 

pollen grain size and seed siring (Cruzan, 1990). Larger pollen grains may have greater 

resource content (e.g., starch granules or lipids) which increases the probability of an ovule’s 

maturation due to increased metabolic vigour (Cruzan, 1990). Notably, diploids produce pollen 

grains of uniform size and shape within a species. In contrast, triploids produce pollen of 

irregular size and shape because individuals exhibit high levels of aneuploidy due to irregular 

meiosis during microsporogenesis (Aparicio, 1994; Tas and van Dijk, 1999; Mráz et al., 2002). 

The size and shape of pollen grains has implications for its germinability. Diploids often have 

higher pollen germination rates than both tetraploids and triploids (Liu et al., 2003; Maggi et 

al., 2008; Ovchinnikov et al., 2017).  

Pollen tube growth rates can be affected by ploidy level (Lankinen et al., 2009). Pollen tube 

growth in diploids is usually faster than in triploids, which may be due to the high proportion 

of aneuploid pollen grains produced by triploids (Lankinen et al., 2009). An increase in DNA 

content results in an increase in cell size, nuclear size, and the duration of the cell cycle. This 

negatively effects pollen tube growth rates as more time is required for pollen tube wall 

formation (e.g., Soares et al., 2014). Pollen tube growth rate is positively correlated with the 

fitness of progeny (Walsh and Charlesworth, 1992). A lower pollen tube growth rate may be a 

pre-zygotic barrier that prevents fertilization from pollen with malfunctioning genomes 

(Lankinen et al., 2009). The visualisation of pollen germinability and pollen tube growth allows 

for the identification of pre-zygotic reproductive barriers (Atlagić et al., 2012). These barriers 

consequently influence the plant breeding strategies that are used to mitigate such 

incompatibilities (Atlagić et al., 2012). 

The reproductive strategy used by a species has implications for the occurrence of interploidy 

mating (Yamauchi et al., 2004). The conjugation of haploid and diploid pollen produced by 

diploid and tetraploid individuals, respectively, may result in the formation of triploids, 

however these triploids typically exhibit lower fitness levels than their progenitors (Burton and 



35 
 

Husband, 2000). This lower fitness can occur via two mechanisms: endosperm collapse due to 

the unbalanced ratio of maternal and paternal genomes known as the ‘triploid block’ or the 

reduced production of gametes due to irregular meiosis (Ramsey and Schemske, 1998). Both 

of these factors may limit the role triploids play in polyploid evolution by preventing them 

from reproducing (Thompson and Lumaret, 1992). Conversely, triploids may facilitate the 

formation of tetraploids via a ‘triploid bridge’ (Ramsey and Schemske, 1998). Triploid bridges 

occur when haploid gametes from diploids fuse with unreduced gametes from triploids thereby 

forming tetraploids (Yahara, 1990). Burton and Husband (2001) suggested that partial viability 

and/or fertility of triploids may facilitate tetraploid establishment and result in the formation of 

genetically diverse progeny. 

Apomixis, clonal reproduction by seed, is a common reproductive strategy used by polyploids 

(Ramsey and Schemske, 1998). This has been observed in Sorbus L. (Robertson et al., 2010), 

Amelanchier Medik (Burgess et al., 2014) Boechera A. Love and D. Love (Luise et al., 2012) 

and Potentilla L. (Dobes et al., 2013) among other genera. Meiosis is often deregulated (the 

reduction or elimination of certain steps in the meiotic process) in newly formed hybrids, which 

leads to increase in the production of unreduced gametes (Asker and Jerling, 1992). This often 

results in individuals favouring apomixis to avoid the formation of lower fitness offspring due 

to genetic abnormalities (Ramsey and Schemske, 1998). The type of apomixis expressed within 

a population can thereafter have implications for pollen morphology and viability (Whitton et 

al., 2008). 

Obligate apomictic plants often have low pollen viability or complete pollen sterility (Maia et 

al., 2015). This is presumably due to individuals occurring in areas with little or no insect 

visitors (Baker, 1967). In contrast, facultative and pseudogamous apomictic plants have high 

pollen viability (Maia et al., 2015). Pseudogamous apomictic plants require high pollen 



36 
 

viability for the fertilisation of the polar nuclei. Due to the mixed breeding system used by 

facultative apomictic plants, high pollen viability is required for sexual reproduction to occur 

successfully (Maia et al., 2015). Consequently, pollen viability is a reliable indicator of the 

type of apomixis used by a plant species, once it is established that apomixis occurs (Maia et 

al., 2015). 

Pollen production constitutes a substantial reproductive cost in plants (Meirmans et al., 2006; 

Mráz, 2009). For example, male-sterile apomictic dandelions produce more flower heads per 

plant, and consequently more seeds, than pollen producing apomictic dandelion species 

(Meirmans et al., 2006). Pollen production may be maintained in phylogenetically recent 

autonomous apomictics because they have not yet accumulated sufficient mutations for male 

sterility (Smith, 1978). As the frequency of apomixis increases however, pollen transmission 

become less efficient and results in pollen fertility declining over time (Smith, 1978). 

Introduced populations of Campuloclinium macrocephalum (Less.) DC. (Asteraceae, 

Eupatorieae) in South Africa have been shown to comprise triploid and tetraploid individuals 

with a few aneuploid individuals among them (Gitonga et al., 2022). The factors contributing 

to the invasiveness of this species are not well understood (McConnachie et al., 2011; Farco 

and Dematteis, 2014) however, polyploidy may be enhancing the species’ potential for habitat 

colonisation via a triploid bridge. In a previous study, Farco and Dematteis (2014) conducted 

a study on the meiotic system and pollen viability (estimated via staining techniques) in C. 

macrocephalum populations in Argentina and Uruguay and found variable meiotic behaviour 

with irregular chromosome pairing in both triploids and tetraploids. Meiotic indices suggested 

that only four out of fourteen South American populations were meiotically stable and were 

therefore normally fertile. The other ten populations had variable pollen viability, with greater 

pollen viability in triploids than tetraploids. The low viability of male gametes and the 

expression of apomixis in C. macrocephalum polyploids in Argentina and Uruguay was 



37 
 

attributed to meiotic abnormalities and subsequently reduced probabilities of producing viable 

offspring. 

Male fertility (i.e., the ability of pollen grains to sire seeds) in populations of C. macrocephalum 

in South Africa has not yet been assessed. An investigation of male fertility is important 

because viable pollen could enable gene flow between populations of different ploidy levels 

(Alexander, 2020). Rapidly evolving populations of C. macrocephalum would be difficult to 

eradicate. Moreover, complete, or partial male sterility could promote reproduction via 

apomixis, a reproductive mode that allows for the rapid production of large numbers of 

offspring (Farco and Dematteis, 2014). In South Africa, C. macrocephalum is hypothesised to 

reproduce via apomixis due to the low genetic variation among populations despite the prolific 

seed production (Gitonga et al., 2015). However, this aspect has not been investigated further. 

Assessing male fertility could provide information about the breeding system strategies used 

by C. macrocephalum, the potential of interploidy mating and its implications for biocontrol. 

The aim of this investigation is to assess male fertility using comparisons of pollen size, shape, 

and viability between selected triploid and tetraploid populations of C. macrocephalum. These 

data will help me to infer possible reproductive barriers in populations of different ploidy 

levels. This study will improve our understanding of the reproductive strategies used by C. 

macrocephalum – a factor which may be contributing to the species invasiveness in South 

Africa. I hypothesize that triploids will exhibit lower male fertility than tetraploids due to 

irregular meiosis. This should reduce gene flow between triploid and tetraploid cytotypes. 

METHOD AND MATERIALS 

 

Study sites 

This investigation was conducted on four populations of C. macrocephalum in the Gauteng 

province of South Africa. These populations were in Tembisa (triploid, 26.05 S; 28.16 E), 



38 
 

Midrand (triploid, 26.02 S; 28.13 E), Modderfontein Nature Reserve (tetraploid, 26.09 S; 28.15 

E) and Greenstone (tetraploid, 26.12 S; 28.15 E). These populations were selected based on 

their putative ploidy levels identified by Gitonga et al. (2022). 

Study species  

Campuloclinium macrocephalum, referred to as the pompom weed, is a South American 

perennial herb that has invaded South African grasslands. The species is conspicuous during 

its flowering months from December to March. The pompom weed is characterised by its pink 

flowerheads that are produced in dense clusters. The leaves form a rosette at the base of the 

plant; however, they decrease in number and size along the length of the stem. The stem and 

leaves are covered in rough hairs. Adult plants grow to approximately 1.5 m tall. At the end of 

the flowering season, mature florets each produce a cypsela with tufts of brown hair. The 

species has a woody rootstock with perennial roots. In summer, new shoots emerge from the 

rootstock and in autumn they die back down. The rootstock, constituting a xylopodium, enables 

the species to persist when its aerial parts die.  

Pollen viability 

Aniline blue stain 

After anthesis, capitula from thirty individuals per population were cut and placed in brown 

paper bags. Five florets from each individual were placed in Eppendorf tubes containing 0.5 

ml of 0.1% aniline blue to preserve them for laboratory analyses. Prior to viewing under a 

microscope, the anthers were dissected out of the florets using a dissecting needle and placed 

back into the vials. The vials were then vortexed for one minute to resuspend the pollen grains. 

Thereafter, 40 µL of the aniline blue solution was placed on a microscope slide and a coverslip 

was placed over it. A 25 mm x 25 mm plastic grid was then placed over the coverslip. The 

slides were examined under a Zeiss compound light stereomicroscope at 400× and the number 



39 
 

of all the viable and non-viable pollen grains in the 40 µL of the aniline blue solution were 

counted.  

Acetocarmine stain 

Pollen viability using the acetocarmine stain was assessed following Farco and Dematteis 

(2014). This stain was used to confirm the pollen viability results from the aniline blue stain as 

the aniline blue stain made it difficult to differentiate between viable and non-viable grains. 

Capitula from 10 individuals per population were harvested and fixed in Carnoy’s fixative 

(6:3:1, 95% ethanol: glacial acetic acid: chloroform) for 2 hours. Thereafter, the capitula were 

stored in 70% ethanol until further analyses. Prior to staining, all the anthers from three florets 

per individual were dissected and placed on a microscope slide with a drop of 70% ethanol to 

prevent the tissue from drying out. The anthers were then squashed to extract the pollen grains 

and the excess anther tissue was removed from the slide. A drop of acetoglycerol stain was 

then added to the slide and a coverslip was placed over it. The slides were then examined under 

Zeiss compound light stereomicroscope at 400x. Dark red stained nuclei were indicative of 

viable pollen grains while unstained/lightly stained nuclei were indicative of non-viable pollen 

grains.  

The percentage viability/non-viability for both stains was calculated as follows: 

Percentage viability =
Number of viable/ non viable pollen grains

Total number of pollen grains
 x 100 

After the mean percentage viability was calculated, the distribution of pollen grain viability 

percentages was tested using a Shapiro Wilk test in R studio (R version 4.0.3) using the fBasics 

package (Wuertz et al., 2017). The data were found to be non-parametric, therefore a Kruskall-

Wallis test was used to test for differences between the pollen viability percentages of the four 

populations. 



40 
 

Pollen tube analyses 

Pollen tube analyses followed the protocol outlined by Kalinganire et al., (2000) for the 

visualisation of pollen tube growth. Twenty florets from ten individuals per population were 

removed from the capitula harvested from “open treatments” (i.e., naturally occurring capitula) 

under a Zeiss Stereo Discovery V12 dissecting microscope. The florets were then fixed in 

‘Carnoy’s’ fixative, comprising ethanol:chloroform:acetic acid (6:3:1), for two hours. 

Thereafter, the florets were transferred to 70% ethanol for storage. The pistils were not softened 

or cleared using NaOH as the delicate tissue disintegrated during this step. The pistils were 

rinsed with distilled water and stained with aniline blue in potassium buffer for two hours; 

longer durations made it difficult to work with the tissue. The florets were then placed on a 

microscope slide with a drop of 80% glycerol and squashed using a coverslip. The slides were 

examined under the BX63 OFM fluorescent microscope to observe pollen tube growth.  

To confirm the results obtained from fluorescence microscopy, five pistils from three 

individuals per population were dissected out of florets and placed on carbon tape adhered to 

aluminium stubs. The pistils were left to dry overnight and then coated with one coat of carbon 

and one coat of gold/palladium. The stubs were viewed under a Tescan Vega Scanning Electron 

Microscope. The presence of pollen tube growth was indicative of the viability of the pollen 

grains. 

Pollen grain size and shape 

The slides prepared using the aniline blue stain were used to measure pollen grain size and 

shape. Five florets from thirty individuals were placed in Eppendorf tubes with the aniline blue 

stain. Thereafter, the tubes were vortexed and 40 µL of the solution was deposited onto a slide 

and covered with a coverslip. The size of 50 viable pollen grains and all of the non-viable grains 

(as there were very few non-viable pollen grains) from 10 individuals per population were 



41 
 

measured using the available function on the Axio Imager connected to a Carl Zeiss compound 

light microscope at 400×. Irregularities in the shape of pollen grains were also noted. Pollen 

shape was described and compared qualitatively between the putative tetraploid and triploid 

populations. The distribution of the pollen grain size data was tested using a Shapiro Wilk test 

on R studio (R version 4.0.3) using the fBasics package (Wuertz et al., 2017). The data were 

analysed for significant differences in pollen grain size between the populations using a 

Kruskal-Wallis test followed by a Kruskal-Wallis multiple comparison post hoc test in R studio 

(R version 4.0.3) using the pgirmess package (Giraudoux et al., 2018). Within population 

variation of pollen grain sizes was also evaluated (mean ± standard deviation).  

Flow cytometry and anther squashes 

Flow cytometry was used to confirm the ploidy levels obtained by Gitonga et al. (2022) in 

2013–2014 for the study populations. Flow cytometry provides estimates of DNA content 

relative to a known standard from which ploidy level can be inferred. The protocol outlined in 

Dolezel et al, (2007) was followed. Seedlings from three maternal plants per population were 

grown in a growth chamber (at 25 ºC and 60% humidity on a 16-hour light schedule) until they 

were large enough to harvest. Leaves from two offspring per maternal plant were used to 

estimate ploidy. Zea mays L. was used as the internal standard. A 1 x 1 cm section of leaf tissue 

from Z. mays was added to a Petri dish containing a 1 x 1 cm section of C. macrocephalum leaf 

tissue. The leaves were simultaneously chopped using a razor in a 500 µl lysis buffer (Otto I). 

Unhomogenised leaf tissue was removed from the solution by filtering it through 30 µm mesh 

filters. Thereafter, the solution was incubated for 15 minutes at room temperature. A 1000 µl 

of Otto II containing DAPI (4’, 6-diamidino-2-phenylindole) was added to the solution to stain 

the cells. This process was repeated for all the samples. The samples were then analysed using 

a CyFlowR Space flow cytometer at the University of Pretoria. FloMax software (Partec, 



42 
 

Münster, Germany) was used to calculate the mean size of each peak (sample and standard) 

and calculate the coefficient of variation. 

Relative DNA content was calculated as follows: 

Relative DNA content = Fluorescence value of the sample/Fluorescence value of the standard 

Anther squashes were used to determine chromosome number and to ascertain if irregular 

meiosis occurs in C. macrocephalum. Anther squashes were conducted following Windham et 

al. (2020). Young capitula were harvested from five individuals per population and fixed in 

Carnoy's fixative (6:3:1 95% ethanol:chloroform:glacial acetic acid) for 24 hours. Thereafter, 

the fixed material was transferred into 70% ethanol and stored in a freezer at -20 ºC. To prepare 

the material for squashing, an array of buds was chosen and placed in a Petri dish with 70% 

ethanol to prevent buds from drying out. The anthers were excised and if they were white and 

showed normal development, they were selected for the staining process. A drop of diluted 

acetocarmine stain was placed on one side of a microscope slide. The anthers were transferred 

to this drop and further isolated from surrounding tissue. A drop of full strength acetocarmine 

stain was placed on the other side of the microscope slide and the anthers were transferred to 

it. A dissecting needle was then used to crush the anthers until the sample could not be 

homogenized anymore. All of the unhomogenized material was then removed and a drop of 

Hoyer’s solution was placed on the slide. The Hoyer’s solutions reduced coverslip rebound and 

partially de-stained the cytoplasm. A coverslip was then lowered onto the droplet and gently 

tapped to get rid of air bubbles. The slide was then placed in paper towel and squashed. The 

slide was viewed under a Carl Zeiss compound light stereomicroscope and photographs were 

taken of the observed stages of meiosis/mitosis on the slide at 400x and/or 1000x. 

RESULTS 

Pollen viability  



43 
 

There was no significant difference between pollen viability percentages amongst the four 

populations using the aniline blue stain (H3 = 7.43; P = 0.06) (Figure 2.1). When stained with 

aniline blue, the average pollen viability of populations of C. macrocephalum was 98%. The 

acetoglycerol stain, however, revealed an average pollen viability of 90% for the same 

populations. There was no significant difference between average pollen viability amongst the 

four populations using the acetoglycerol stain (H3 = 0.64; P = 0.59, Figure 2.1). The 

discrepancies between the data obtained from the two stains may be attributed to the difficulty 

in differentiating between viable and non-viable pollen grains using the aniline blue stain 

(Figure 2.2). 

Figure 2.1. Mean pollen viability (%) in four populations of Campuloclinium macrocephalum 

collected across Gauteng, South Africa. Thirty plants per population were assessed using 

aniline blue stain and ten plants per population were assessed using the acetoglycerol stain. The 

bars above the column represent standard error. The letters above each bar represent statistical 

significance between populations for each stain. 

 

 

 



44 
 

 

Figure 2.2. Pollen grains of Campuloclinium macrocephalum stained for viability assessments 

using acetocarmine (A, B, C) and acetoglycerol (D, E, F). A: Pollen grain stained by 

acetoglycerol with an arrow indicating a non-viable pollen grain (scale bar = 24 µm), B: viable 

pollen grain stained by acetoglycerol (scale bar = 18 µm), C: non-viable pollen grain stained 

by acetoglycerol (scale bar = 13 µm), D: Pollen grains stained by aniline blue with an arrow 

indicating a non-viable pollen grain (scale bar = 22 µm), E: viable pollen grain stained by 

aniline blue (scale bar = 15 µm), F: non-viable grain stained by aniline blue (scale bar = 17 

µm).  

 

Pollen tube analyses  

Pollen tube analyses were also used as an indicator of male fertility. No pollen tube growth 

was observed using fluorescence microscopy, however scanning electron microscopy showed 

pollen tube growth on the outside of the style and on the stigmatic margin (Figure 2.3). The 

location on the style where the pollen grains adhered matched where the anthers enclose the 

style before the latter emerges thus suggesting that it is autogamous pollen (Figure 2.4). The 

presence of pollen tube growth on both sides of the style branches, including near the stigmatic 

surface, indicates that the species produces viable pollen and therefore supports the pollen 

viability count data. 



45 
 

 

 

Figure 2.3. Pollen tube growth in Campuloclinium macrocephalum observed under a scanning 

electron microscope. A–B: Pollen grains on the style with no visible pollen tube growth (scale 

bar in A = 100µm, B = 50 µm), C–D: Pollen germinating on the stigmatic margin (scale bar in 

C = 20 µm, D = 50 µm) and E–F: Pollen grains with pollen tube growth situated on the outside 



46 
 

of the styles (scale bar in E = 10 µm, F = 5 µm). ST: style, SM: stigmatic margin and PT: pollen 

tube. 

 

Figure 2.4. The position of the anthers relative to the style in developing florets of 

Campuloclinium macrocephalum. A) A ring of fused anthers surrounding the style branches, 

with appressed  style tips, before it elongates, B) Elongated style branches with separated style 

tips. Scale bars = 1.7mm. 

 

Pollen grain diameter 

There is a significant difference between viable and non-viable pollen grain diameter in all four 

populations (H7 = 53.81; P < 0.0001, Table 2.1). Viable pollen grain diameter also differs 

significantly amongst the four populations (H3
 = 204.9; P< 0.0001, Figure 2.5). There was a 

significant difference between Tembisa (3x) and Midrand (3x) (P < 0.001), Tembisa and 

Modderfontein (4x) (P < 0.001), Tembisa and Greenstone (4x) (P < 0.001), and Modderfontein 

and Midrand (P < 0.001). There was no significant difference between Greenstone and Midrand 

(P = 0.311), Greenstone and Modderfontein (P = 0.133) (Appendix 2.1). 

 



47 
 

Table 2.1. Mean pollen grain diameter (± SE) of viable and non-viable pollen grains in four 

populations of Campuloclinium macrocephalum. 

Population and putative 

ploidy level 

Mean diameter of viable 

pollen grains (±SE) 

Mean diameter of non-

viable pollen grains (±SE) 

Tembisa (3x) 24.05 ± 0.70 21.19 ± 0.82 

Midrand (3x) 27.92 ± 0.71 25.58 ± 0.62 

Modderfontein (4x) 26.43 ± 0.34 21.62 ± 0.30 

Greenstone (4x) 26.98 ± 0.30 23.52 ± 0.41 

 

 

 

Figure 2.5. Viable pollen grain diameter in four South African populations of Campuloclinium 

macrocephalum using aniline blue. Tembisa and Midrand are triploid populations while 

Modderfontein and Greenstone are tetraploid populations. The circles represent outliers, the 

bars represent minimum and maximum values and the box represents variation in the data and 

the midline represents the mean. 

 

 

 

 



48 
 

Pollen grain morphology 

Pollen grains of C. macrocephalum are spherical and echinate (i.e., covered in micro-spines). 

The spines have a regular arrangement that was consistent in all of the observed pollen grains. 

Pollen grain tetrads and triads were found in all sampled individuals from each population. The 

presence of triads is indicative of irregular meiosis occurring in this species. Most of the tetrads 

and triads were stained as viable by the aniline blue stain, however they were stained as non-

viable using the acetoglycerol stain (Figure 2.6). While pollen grain shape was consistent in all 

four populations, pollen grain size differed (Figure 2.6). 

Flow cytometry and anther squashes  

From the 24 individuals analysed using flow cytometry, four samples had to be excluded 

because their coefficient of variation exceeded 5%, which does not provide a reliable 

estimation of DNA content. The estimated ploidy for the remaining 20 individuals was based 

on their relative 2C DNA content (Appendix 2.2; Appendix 2.3). Relative 2C DNA content 

ranged from 2.87 – 3.26, indicating that they were most likely triploid individuals. This was 

corroborated by anther squashes revealing a chromosome number of approximately 15 (base 

chromosome number x = 10; Farco et al. (2012)). It was challenging to obtain exact counts due 

to the limitations of the technique (such as obtaining anthers with a high meiotic index) and the 

available equipment (Figure 2.7). All of the offspring from the Modderfontein and Greenstone 

populations were identified as triploid despite being harvested from populations previously 

identified as tetraploid. 

 



49 
 

Figure 2.6. Size variation in pollen grains of Campuloclinium macrocephalum. A) Viable triad 

grain indicated by the arrow (scale bar = 28 µm), B) viable tetrad pollen grain (scale bar = 10 

µm), C) viable pollen grains with consistent shape (scale bar = 10 µm), D) non-viable tetrad 

pollen grain (scale bar = 28 µm), E) non-viable triad pollen grains indicated by the arrow (scale 

bar = 27 µm) and F) viable pollen grains with consistent shape but different sizes (scale bar = 

30 µm). Pollen grains in panels A–C were stained by aniline blue while pollen grains in panels 

D–F were stained by acetocarmine.  



50 
 

Figure 2.7. Meiotically dividing cells obtained from anther squashes. Both panels show cells 

in interphase and arrows indicate chromosomes. Scale bars = 18 µm.  

 

DISCUSSION  

 

The purpose of this study was to assess male fertility in South African populations of 

Campuloclinium macrocephalum by estimating pollen viability using staining techniques, 

pollen tube analyses, and pollen grain characterisation. Additionally, flow cytometry and 

anther squashes were used to determine the putative ploidy level of offspring formed in each 

population. Collectively, these data were used to determine the relationship between male 

fertility and ploidy level thereby enabling inferences on the potential gene flow between 

populations and its implications for biocontrol management. 

The relative fitness of pollen grains is determined by estimating their viability and fertility 

(Wizenberg et al., 2021). Pollen grain viability, estimated by staining techniques, measures the 

number of pollen grains that can engage in reproduction under any conditions (Alexander, 

1969; Wizenberg et al., 2021) while male fertility is measured by determining the number of 

pollen grains that can germinate u